Paddock
Per-task discrimination
A task discriminates when at least one scored attempt passed it and at least one did not. A task every combination passes, or none does, cannot rank anything; it is a candidate for a harder version.
coding
| task | version | passed / n | combos passing | discriminates | runs (P passed, F not passed) |
|---|---|---|---|---|---|
| coding-cache-context | 1 | 58/62 | 57/61 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-opus-5-5 @low: P; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| coding-commit-ack | 1 | 54/60 | 53/59 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-opus-5-5 @low: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: F; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode local/qwen3-coder-next-80b-a3b-6bit: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| coding-generation-fence | 1 | 57/62 | 56/61 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: F; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6-sol @xhigh: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| coding-jsonb-like-predicate | 1 | 82/86 | 59/61 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: PF; claude-code fireworks/kimi-k3: PP; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: P; claude-code moonshot/kimi-k3: PP; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PP; claude-code zai/glm-5.2: PP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: PP; codex fireworks/kimi-k3: PP; codex fireworks/qwen3.8-max: PP; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: PP; codex together/deepseek-v4.1-flash: PP; codex zai/glm-5.2: PP; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; codex moonshot/kimi-k3: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PP; opencode fireworks/kimi-k3: PP; opencode fireworks/qwen3.8-max: PP; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: P; opencode moonshot/kimi-k3: PP; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PPP; opencode zai/glm-5.2: PP; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FP; pi fireworks/kimi-k3: PP; pi fireworks/qwen3.8-max: PP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: PP; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PP; pi zai/glm-5.2: PP |
| coding-missing-commit | 1 | 84/89 | 61/64 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: PP; claude-code baseten/glm-5.3-flash: PP; claude-code fireworks/kimi-k3: PP; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: P; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PP; claude-code zai/glm-5.2: PP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: PP; codex fireworks/kimi-k3: PP; codex fireworks/qwen3.8-max: PP; codex google/gemini-3.8-flash: F; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: PP; codex together/deepseek-v4.1-flash: PF; codex zai/glm-5.2: PP; codex openai/gpt-5.6-luna @low: P; codex openai/gpt-5.6-sol @low: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; codex moonshot/kimi-k3: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PP; opencode fireworks/kimi-k3: PP; opencode fireworks/qwen3.8-max: PP; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: P; opencode moonshot/kimi-k3: PP; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PPP; opencode zai/glm-5.2: PP; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PF; pi fireworks/kimi-k3: PP; pi fireworks/qwen3.8-max: PP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: P; pi moonshot/kimi-k3: PP; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PP; pi zai/glm-5.2: PP |
| coding-nondict-migrate-audit | 1 | 79/83 | 57/60 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: PP; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: P; claude-code moonshot/kimi-k3: PP; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PP; claude-code zai/glm-5.2: PP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: PP; codex fireworks/kimi-k3: PP; codex fireworks/qwen3.8-max: PP; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: PP; codex zai/glm-5.2: PP; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; codex moonshot/kimi-k3: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: FP; opencode fireworks/kimi-k3: PP; opencode fireworks/qwen3.8-max: PP; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: F; opencode moonshot/kimi-k3: PP; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PPP; opencode zai/glm-5.2: PP; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PP; pi fireworks/kimi-k3: PP; pi fireworks/qwen3.8-max: PP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: PP; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PP; pi zai/glm-5.2: PP |
| coding-pipefail-silent-exit | 1 | 125/141 | 57/64 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: PP; claude-code fireworks/kimi-k3: PP; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: F; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPF; claude-code zai/glm-5.2: PP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: PP; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: FP; codex zai/glm-5.2: PPP; codex openai/gpt-6-astra @low: P; codex openai/gpt-6-astra @xhigh: P; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; codex moonshot/kimi-k3: F; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PFP; opencode fireworks/kimi-k3: FF; opencode fireworks/qwen3.8-max: PP; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: F; opencode moonshot/kimi-k3: PP; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: PPF; opencode zai/glm-5.2: FP; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPF; pi fireworks/kimi-k3: FP; pi fireworks/qwen3.8-max: PPF; pi google/gemini-3.8-flash: PPPPP; pi google/gemini-3.8-flash @high: PPPPP; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: PP; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: PP |
| coding-pr-not-found-terminal | 1 | 111/143 | 56/63 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: FP; claude-code fireworks/kimi-k3: FP; claude-code fireworks/qwen3.8-max: FPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: P; claude-code moonshot/kimi-k3: FP; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: FPP; claude-code zai/glm-5.2: FP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: PP; codex fireworks/kimi-k3: FPP; codex fireworks/qwen3.8-max: FP; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: F; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: FP; codex together/deepseek-v4.1-flash: FP; codex zai/glm-5.2: FPF; codex openai/gpt-6-sol @medium: P; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; codex moonshot/kimi-k3: F; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: FPP; opencode fireworks/kimi-k3: FP; opencode fireworks/qwen3.8-max: FP; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: F; opencode moonshot/kimi-k3: FP; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: FPF; opencode zai/glm-5.2: FP; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FPP; pi fireworks/kimi-k3: FP; pi fireworks/qwen3.8-max: FPP; pi google/gemini-3.8-flash: PPPPP; pi google/gemini-3.8-flash @high: PPPPP; pi google/gemini-3.8-flash @low: F; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: FP; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: FPPPP; pi zai/glm-5.2: FP |
| coding-registry-tail | 1 | 57/60 | 56/59 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6-sol @medium: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| coding-schema-behind-wording | 1 | 117/143 | 50/63 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: F; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5: F; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: FP; claude-code fireworks/kimi-k3: PP; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: F; claude-code local/qwen3-coder-next-80b-a3b-6bit: F; claude-code moonshot/kimi-k3: PP; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: FPP; claude-code zai/glm-5.2: PP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: PP; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: PP; codex google/gemini-3.8-flash: F; codex google/gemini-3.8-flash @high: F; codex google/gemini-3.8-flash @low: F; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: PP; codex together/deepseek-v4.1-flash: PP; codex zai/glm-5.2: PPF; codex openai/gpt-6-astra @low: P; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; codex moonshot/kimi-k3: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: PP; opencode fireworks/qwen3.8-max: PP; opencode google/gemini-3.8-flash: F; opencode google/gemini-3.8-flash @high: F; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: P; opencode moonshot/kimi-k3: PP; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: PPP; opencode zai/glm-5.2: PP; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: PP; pi fireworks/qwen3.8-max: PFP; pi google/gemini-3.8-flash: PFFFF; pi google/gemini-3.8-flash @high: FFPFF; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: PP; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: PF |
| coding-toc-commands | 1 | 58/62 | 57/61 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: F; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: P; codex openai/gpt-5.6-luna @medium: P; codex openai/gpt-5.6-terra @high: P; codex openai/gpt-6-luna @high: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: F; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| coding-toc-large-output | 1 | 97/109 | 45/49 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: PPPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PFPPP; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPP; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: FFP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: FPP; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: FPF; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPPPP; opencode together/deepseek-v4.1-flash: F; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPF; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: PPP; pi together/deepseek-v4.1-flash: PPPPF; pi zai/glm-5.2: P |
| coding-toc-large-output | 2 | 116/124 | 55/59 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: P; claude-code anthropic/claude-opus-5-5: PPPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: F; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: PFP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: FPP; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: FPF; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPPPP; opencode together/deepseek-v4.1-flash: F; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: PPPP; pi google/gemini-3.8-flash @high: PPP; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: PPP; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: P |
| coding-unknown-change-kind-sentinel | 1 | 93/138 | 46/61 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: FF; claude-code fireworks/kimi-k3: FP; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code local/qwen3-coder-next-80b-a3b-6bit: P; claude-code moonshot/kimi-k3: FP; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: FPF; claude-code zai/glm-5.2: FP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: FF; codex fireworks/kimi-k3: FFF; codex fireworks/qwen3.8-max: FP; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex local/qwen3-coder-next-80b-a3b-6bit: F; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: FF; codex zai/glm-5.2: FPP; codex openai/gpt-6.1-sol: FPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPFPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; codex moonshot/kimi-k3: F; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: FFF; opencode fireworks/kimi-k3: FP; opencode fireworks/qwen3.8-max: FF; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode local/qwen3-coder-next-80b-a3b-6bit: P; opencode moonshot/kimi-k3: FF; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: FFF; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FPF; pi fireworks/kimi-k3: FP; pi fireworks/qwen3.8-max: FPP; pi google/gemini-3.8-flash: PPPP; pi google/gemini-3.8-flash @high: PPPPP; pi google/gemini-3.8-flash @low: P; pi local/qwen3-coder-next-80b-a3b-6bit: P; pi moonshot/kimi-k3: FP; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: FPFPF; pi zai/glm-5.2: FP |
docs
| task | version | passed / n | combos passing | discriminates | runs (P passed, F not passed) |
|---|---|---|---|---|---|
| docs-cache-config | 1 | 21/47 | 18/36 | yes | antigravity google/gemini-3.1-pro @high: PPP; antigravity google/gemini-3.8-flash @high: P; claude-code anthropic/claude-haiku-4-5: F; claude-code anthropic/claude-opus-5-5: FF; claude-code anthropic/claude-sonnet-5: FF; claude-code anthropic/claude-sonnet-5-5: PF; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: FFF; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: FF; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: F; codex moonshot/kimi-k3: F; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: PPF; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode together/deepseek-v4.1-flash: FP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: F; pi google/gemini-3.8-flash: P; pi moonshot/kimi-k3: F; pi zai/glm-5.2: P |
| docs-cache-config | 2 | 51/53 | 51/53 | yes | antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| docs-config-loader-keys | 1 | 11/47 | 11/36 | yes | antigravity google/gemini-3.1-pro @high: FFF; antigravity google/gemini-3.8-flash @high: P; claude-code anthropic/claude-haiku-4-5: F; claude-code anthropic/claude-opus-5-5: PF; claude-code anthropic/claude-sonnet-5: FF; claude-code anthropic/claude-sonnet-5-5: PF; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: FFP; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: FF; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: F; codex fireworks/qwen3.8-max: F; codex moonshot/kimi-k3: F; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: FFF; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: F; opencode together/deepseek-v4.1-flash: FF; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: F; pi google/gemini-3.8-flash: P; pi moonshot/kimi-k3: F; pi zai/glm-5.2: P |
| docs-config-loader-keys | 2 | 53/53 | 53/53 | no | antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| docs-gym-baseline-keys | 1 | 13/48 | 11/37 | yes | antigravity google/gemini-3.1-pro @high: FFF; antigravity google/gemini-3.8-flash @high: F; claude-code anthropic/claude-haiku-4-5: F; claude-code anthropic/claude-opus-5-5: FF; claude-code anthropic/claude-sonnet-5: FF; claude-code anthropic/claude-sonnet-5-5: PF; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPF; claude-code moonshot/kimi-k3: F; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: PP; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: F; codex moonshot/kimi-k3: P; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: FPF; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: F; opencode together/deepseek-v4.1-flash: FF; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: F; pi google/gemini-3.8-flash: F; pi moonshot/kimi-k3: F; pi zai/glm-5.2: F |
| docs-gym-baseline-keys | 2 | 27/39 | 27/39 | yes | antigravity google/gemini-3.8-flash @high: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: F; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: F; opencode together/deepseek-v4.1-flash: F; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: F; pi moonshot/kimi-k3: F; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| docs-gym-baseline-keys | 3 | 35/53 | 35/53 | yes | antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: F; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: F; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: F; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: F; opencode google/gemini-3.8-flash @high: F; opencode google/gemini-3.8-flash @low: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: F; opencode together/deepseek-v4.1-flash: F; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: F; pi google/gemini-3.8-flash @high: F; pi google/gemini-3.8-flash @low: F; pi moonshot/kimi-k3: F; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| docs-gym-run-flags | 1 | 2/47 | 2/37 | yes | antigravity google/gemini-3.1-pro @high: FFF; antigravity google/gemini-3.8-flash @high: F; claude-code anthropic/claude-haiku-4-5: F; claude-code anthropic/claude-opus-5-5: FP; claude-code anthropic/claude-sonnet-5: FF; claude-code anthropic/claude-sonnet-5-5: FF; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: FFF; claude-code moonshot/kimi-k3: F; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: FF; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: F; codex fireworks/qwen3.8-max: F; codex moonshot/kimi-k3: F; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: FF; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: F; opencode together/deepseek-v4.1-flash: FF; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: F; pi google/gemini-3.8-flash: F; pi moonshot/kimi-k3: F; pi zai/glm-5.2: F |
| docs-gym-run-flags | 2 | 44/53 | 44/53 | yes | antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: F; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: F; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: F; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: F; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: F; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: F |
| docs-retry-behavior-diff | 1 | 9/46 | 8/36 | yes | antigravity google/gemini-3.1-pro @high: FFF; antigravity google/gemini-3.8-flash @high: P; claude-code anthropic/claude-haiku-4-5: F; claude-code anthropic/claude-opus-5-5: PF; claude-code anthropic/claude-sonnet-5: FF; claude-code anthropic/claude-sonnet-5-5: FP; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PFP; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: FF; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: F; codex fireworks/qwen3.8-max: F; codex moonshot/kimi-k3: F; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: FF; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: F; opencode together/deepseek-v4.1-flash: FF; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: F; pi google/gemini-3.8-flash: P; pi moonshot/kimi-k3: F; pi zai/glm-5.2: P |
| docs-retry-behavior-diff | 2 | 33/39 | 33/39 | yes | antigravity google/gemini-3.8-flash @high: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: F; claude-code google/gemini-3.8-flash: F; claude-code moonshot/kimi-k3: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: F; pi zai/glm-5.2: P |
| docs-retry-behavior-diff | 3 | 51/53 | 51/53 | yes | antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: F; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: F; pi zai/glm-5.2: P |
| docs-snapshot-new-command | 1 | 4/47 | 3/36 | yes | antigravity google/gemini-3.1-pro @high: FFF; antigravity google/gemini-3.8-flash @high: F; claude-code anthropic/claude-haiku-4-5: F; claude-code anthropic/claude-opus-5-5: FFF; claude-code anthropic/claude-sonnet-5: FF; claude-code anthropic/claude-sonnet-5-5: FF; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PFP; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: FF; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: F; codex fireworks/qwen3.8-max: F; codex moonshot/kimi-k3: F; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: FF; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: F; opencode together/deepseek-v4.1-flash: FF; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: F; pi google/gemini-3.8-flash: P; pi moonshot/kimi-k3: P; pi zai/glm-5.2: F |
| docs-snapshot-new-command | 2 | 25/42 | 25/40 | yes | antigravity google/gemini-3.1-pro @high: FFF; antigravity google/gemini-3.8-flash @high: F; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: F; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: F; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi moonshot/kimi-k3: F; pi openai/gpt-6.1-sol: F; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| docs-snapshot-new-command | 3 | 47/56 | 47/54 | yes | antigravity google/gemini-3.1-pro @high: FFF; antigravity google/gemini-3.8-flash @high: F; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: F; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: F; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: F; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
fix
| task | version | passed / n | combos passing | discriminates | runs (P passed, F not passed) |
|---|---|---|---|---|---|
| fix-corrupt-refresh-checkpoint | 1 | 9/11 | 7/9 | yes | antigravity google/gemini-3.1-pro @high: P; claude-code fireworks/qwen3.8-max: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: PP; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: F; pi zai/glm-5.2: F |
| fix-corrupt-refresh-checkpoint | 2 | 88/109 | 42/56 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPFF; claude-code anthropic/claude-opus-5-5 @xhigh: PFFFP; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: F; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PP; claude-code zai/glm-5.2: F; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: F; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: FF; codex openai/gpt-6.1-sol: PPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: F; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: PPP; pi google/gemini-3.8-flash @high: PPPP; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: FPPP; pi zai/glm-5.2: F |
| fix-nested-template-copy | 1 | 55/55 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| fix-publish-guard-order | 1 | 55/55 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| fix-rescue-local-tip | 1 | 12/12 | 11/11 | no | antigravity google/gemini-3.1-pro @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code fireworks/qwen3.8-max: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi zai/glm-5.2: P |
| fix-rescue-local-tip | 2 | 53/53 | 53/53 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| fix-retry-empty-page | 1 | 55/55 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| fix-stale-lock-recovery | 1 | 12/12 | 11/11 | no | antigravity google/gemini-3.1-pro @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code fireworks/qwen3.8-max: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi zai/glm-5.2: P |
| fix-stale-lock-recovery | 2 | 53/54 | 53/54 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
hard-fix
| task | version | passed / n | combos passing | discriminates | runs (P passed, F not passed) |
|---|---|---|---|---|---|
| hard-fix-cross-file | 1 | 55/55 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-cross-file-v2 | 1 | 54/54 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-cross-file-v3 | 1 | 9/9 | 9/9 | no | antigravity google/gemini-4-argon @high: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; codex google/gemini-3.8-flash @low: P; codex openai/gpt-6.1-sol @medium: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @low: P |
| hard-fix-regression-test-dedupe | 1 | 54/54 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-regression-test-dedupe-v2 | 1 | 54/54 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-regression-test-dedupe-v3 | 1 | 1/9 | 1/9 | yes | antigravity google/gemini-4-argon @high: F; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: F; codex google/gemini-3.8-flash @low: F; codex openai/gpt-6.1-sol @medium: P; opencode google/gemini-3.8-flash @high: F; opencode google/gemini-3.8-flash @low: F; pi google/gemini-3.8-flash @low: F |
| hard-fix-regression-test-quiet-hours | 1 | 54/55 | 53/54 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-regression-test-quiet-hours-v2 | 1 | 53/54 | 53/54 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-regression-test-quiet-hours-v3 | 1 | 9/9 | 9/9 | no | antigravity google/gemini-4-argon @high: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; codex google/gemini-3.8-flash @low: P; codex openai/gpt-6.1-sol @medium: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @low: P |
| hard-fix-two-findings | 1 | 55/55 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-two-findings-v2 | 1 | 53/54 | 53/54 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-two-findings-v3 | 1 | 9/9 | 9/9 | no | antigravity google/gemini-4-argon @high: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; codex google/gemini-3.8-flash @low: P; codex openai/gpt-6.1-sol @medium: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @low: P |
| hard-fix-wrong-recommendation-config-reload | 1 | 77/116 | 32/56 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: F; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PPF; claude-code google/gemini-3.8-flash: F; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PFP; claude-code zai/glm-5.2: F; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: FFF; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: F; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: FFF; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: FFF; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode moonshot/kimi-k3: F; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: FF; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FFF; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: PPP; pi google/gemini-3.8-flash @high: PFP; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: F; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPFFF; pi zai/glm-5.2: F |
| hard-fix-wrong-recommendation-retry-charge | 1 | 55/55 | 54/54 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-wrong-recommendation-retry-charge-v2 | 1 | 51/54 | 51/54 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: F; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-fix-wrong-recommendation-retry-charge-v3 | 1 | 4/9 | 4/9 | yes | antigravity google/gemini-4-argon @high: P; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: F; codex google/gemini-3.8-flash @low: F; codex openai/gpt-6.1-sol @medium: P; opencode google/gemini-3.8-flash @high: F; opencode google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @low: F |
hard-review
| task | version | passed / n | combos passing | discriminates | runs (P passed, F not passed) |
|---|---|---|---|---|---|
| hard-review-clean-long | 1 | 69/100 | 36/47 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: FFFFF; claude-code anthropic/claude-opus-5-5 @xhigh: PFPFP; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: FFP; claude-code google/gemini-3.8-flash: F; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PFP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: FPF; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: PPP; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PFFFPPFFPP; codex openai/gpt-6.1-sol @medium: PFPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: PFP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PFF; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PP; pi zai/glm-5.2: P |
| hard-review-clean-long | 2 | 140/218 | 92/158 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: P; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: P; antigravity google/gemini-4-argon @low: P; antigravity google/gemini-4-argon @medium: P; claude-code anthropic/claude-haiku-5-5 @high: F; claude-code anthropic/claude-haiku-5-5 @low: F; claude-code anthropic/claude-haiku-5-5 @max: F; claude-code anthropic/claude-haiku-5-5 @medium: F; claude-code anthropic/claude-haiku-5-5 @xhigh: F; claude-code anthropic/claude-opus-5-5: FFFFF; claude-code anthropic/claude-opus-5-5 @high: F; claude-code anthropic/claude-opus-5-5 @low: F; claude-code anthropic/claude-opus-5-5 @max: P; claude-code anthropic/claude-opus-5-5 @medium: F; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: F; claude-code anthropic/claude-sonnet-5-5 @high: F; claude-code anthropic/claude-sonnet-5-5 @low: F; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: F; claude-code anthropic/claude-sonnet-5-5 @xhigh: F; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: FFP; claude-code google/gemini-3.8-flash: F; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-5.6-terra @high: P; claude-code openai/gpt-5.6-terra @low: F; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: F; claude-code openai/gpt-6-luna @max: P; claude-code openai/gpt-6-luna @xhigh: P; claude-code openai/gpt-6.1-sol: P; claude-code openai/gpt-6.1-sol @high: P; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: P; claude-code together/deepseek-v4.1-flash: PFP; claude-code zai/glm-5.2: P; codex anthropic/claude-haiku-5-5 @high: F; codex anthropic/claude-haiku-5-5 @low: F; codex anthropic/claude-haiku-5-5 @max: F; codex anthropic/claude-haiku-5-5 @medium: F; codex anthropic/claude-haiku-5-5 @xhigh: F; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-opus-5-5 @high: F; codex anthropic/claude-opus-5-5 @low: F; codex anthropic/claude-opus-5-5 @max: P; codex anthropic/claude-opus-5-5 @medium: F; codex anthropic/claude-opus-5-5 @xhigh: P; codex anthropic/claude-sonnet-5-5: F; codex anthropic/claude-sonnet-5-5 @high: F; codex anthropic/claude-sonnet-5-5 @low: F; codex anthropic/claude-sonnet-5-5 @medium: F; codex anthropic/claude-sonnet-5-5 @xhigh: F; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: FPF; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex google/gemini-3.8-flash @medium: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: PPP; codex openai/gpt-5.6-terra @high: F; codex openai/gpt-5.6-terra @low: F; codex openai/gpt-5.6-terra @max: P; codex openai/gpt-5.6-terra @medium: F; codex openai/gpt-5.6-terra @ultra: F; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6-luna @high: F; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: P; codex openai/gpt-6-luna @medium: F; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPFPPP; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: P; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @ultra: P; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-haiku-5-5 @high: P; opencode anthropic/claude-haiku-5-5 @low: P; opencode anthropic/claude-haiku-5-5 @max: F; opencode anthropic/claude-haiku-5-5 @medium: P; opencode anthropic/claude-haiku-5-5 @xhigh: F; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-opus-5-5 @high: F; opencode anthropic/claude-opus-5-5 @low: F; opencode anthropic/claude-opus-5-5 @medium: F; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: F; opencode anthropic/claude-sonnet-5-5 @high: F; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: F; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-5.6-terra @high: F; opencode openai/gpt-5.6-terra @low: P; opencode openai/gpt-6-luna @high: P; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: P; opencode openai/gpt-6-luna @xhigh: P; opencode openai/gpt-6.1-sol: PPP; opencode openai/gpt-6.1-sol @high: P; opencode openai/gpt-6.1-sol @low: P; opencode openai/gpt-6.1-sol @max: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-haiku-5-5 @high: F; pi anthropic/claude-haiku-5-5 @low: F; pi anthropic/claude-haiku-5-5 @max: F; pi anthropic/claude-haiku-5-5 @medium: F; pi anthropic/claude-haiku-5-5 @xhigh: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-opus-5-5 @high: F; pi anthropic/claude-opus-5-5 @low: F; pi anthropic/claude-opus-5-5 @medium: F; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: F; pi anthropic/claude-sonnet-5-5 @high: F; pi anthropic/claude-sonnet-5-5 @low: F; pi anthropic/claude-sonnet-5-5 @medium: F; pi anthropic/claude-sonnet-5-5 @xhigh: F; pi baseten/glm-5.3-flash: PFP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PFP; pi google/gemini-3.8-flash: PP; pi google/gemini-3.8-flash @high: PPP; pi google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @medium: P; pi moonshot/kimi-k3: P; pi openai/gpt-5.6-terra @high: F; pi openai/gpt-5.6-terra @low: F; pi openai/gpt-6-luna @high: P; pi openai/gpt-6-luna @low: F; pi openai/gpt-6-luna @max: P; pi openai/gpt-6-luna @xhigh: P; pi openai/gpt-6.1-sol: P; pi openai/gpt-6.1-sol @high: P; pi openai/gpt-6.1-sol @low: P; pi openai/gpt-6.1-sol @max: P; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: P |
| hard-review-cross-file | 1 | 48/48 | 43/43 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi zai/glm-5.2: P |
| hard-review-cross-file | 2 | 163/164 | 158/159 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: P; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: P; antigravity google/gemini-4-argon @low: P; antigravity google/gemini-4-argon @medium: P; claude-code anthropic/claude-haiku-5-5 @high: P; claude-code anthropic/claude-haiku-5-5 @low: P; claude-code anthropic/claude-haiku-5-5 @max: P; claude-code anthropic/claude-haiku-5-5 @medium: P; claude-code anthropic/claude-haiku-5-5 @xhigh: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-opus-5-5 @high: P; claude-code anthropic/claude-opus-5-5 @low: P; claude-code anthropic/claude-opus-5-5 @max: P; claude-code anthropic/claude-opus-5-5 @medium: P; claude-code anthropic/claude-opus-5-5 @xhigh: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code anthropic/claude-sonnet-5-5 @high: P; claude-code anthropic/claude-sonnet-5-5 @low: P; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: P; claude-code anthropic/claude-sonnet-5-5 @xhigh: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-5.6-terra @high: P; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: P; claude-code openai/gpt-6-luna @max: P; claude-code openai/gpt-6-luna @xhigh: P; claude-code openai/gpt-6.1-sol: P; claude-code openai/gpt-6.1-sol @high: P; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-haiku-5-5 @high: P; codex anthropic/claude-haiku-5-5 @low: P; codex anthropic/claude-haiku-5-5 @max: P; codex anthropic/claude-haiku-5-5 @medium: P; codex anthropic/claude-haiku-5-5 @xhigh: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-opus-5-5 @high: P; codex anthropic/claude-opus-5-5 @low: P; codex anthropic/claude-opus-5-5 @max: P; codex anthropic/claude-opus-5-5 @medium: P; codex anthropic/claude-opus-5-5 @xhigh: P; codex anthropic/claude-sonnet-5-5: P; codex anthropic/claude-sonnet-5-5 @high: P; codex anthropic/claude-sonnet-5-5 @low: P; codex anthropic/claude-sonnet-5-5 @max: P; codex anthropic/claude-sonnet-5-5 @medium: P; codex anthropic/claude-sonnet-5-5 @xhigh: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex google/gemini-3.8-flash @medium: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-5.6-terra @high: P; codex openai/gpt-5.6-terra @low: P; codex openai/gpt-5.6-terra @max: P; codex openai/gpt-5.6-terra @medium: P; codex openai/gpt-5.6-terra @ultra: P; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6-luna @high: P; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: P; codex openai/gpt-6-luna @medium: P; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: P; codex openai/gpt-6.1-sol @medium: P; codex openai/gpt-6.1-sol @ultra: P; codex openai/gpt-6.1-sol @xhigh: P; opencode anthropic/claude-haiku-5-5 @high: P; opencode anthropic/claude-haiku-5-5 @low: P; opencode anthropic/claude-haiku-5-5 @max: P; opencode anthropic/claude-haiku-5-5 @medium: P; opencode anthropic/claude-haiku-5-5 @xhigh: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-opus-5-5 @high: P; opencode anthropic/claude-opus-5-5 @low: P; opencode anthropic/claude-opus-5-5 @medium: P; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: P; opencode anthropic/claude-sonnet-5-5 @high: P; opencode anthropic/claude-sonnet-5-5 @low: P; opencode anthropic/claude-sonnet-5-5 @medium: P; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-5.6-terra @high: P; opencode openai/gpt-5.6-terra @low: P; opencode openai/gpt-6-luna @high: P; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: P; opencode openai/gpt-6-luna @xhigh: P; opencode openai/gpt-6.1-sol: P; opencode openai/gpt-6.1-sol @high: P; opencode openai/gpt-6.1-sol @low: P; opencode openai/gpt-6.1-sol @max: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-haiku-5-5 @high: P; pi anthropic/claude-haiku-5-5 @low: P; pi anthropic/claude-haiku-5-5 @max: P; pi anthropic/claude-haiku-5-5 @medium: P; pi anthropic/claude-haiku-5-5 @xhigh: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-opus-5-5 @high: P; pi anthropic/claude-opus-5-5 @low: P; pi anthropic/claude-opus-5-5 @medium: P; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: P; pi anthropic/claude-sonnet-5-5 @high: P; pi anthropic/claude-sonnet-5-5 @low: P; pi anthropic/claude-sonnet-5-5 @medium: P; pi anthropic/claude-sonnet-5-5 @xhigh: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @medium: P; pi moonshot/kimi-k3: P; pi openai/gpt-5.6-terra @high: P; pi openai/gpt-5.6-terra @low: P; pi openai/gpt-6-luna @high: P; pi openai/gpt-6-luna @low: P; pi openai/gpt-6-luna @max: P; pi openai/gpt-6-luna @xhigh: P; pi openai/gpt-6.1-sol: P; pi openai/gpt-6.1-sol @high: P; pi openai/gpt-6.1-sol @low: P; pi openai/gpt-6.1-sol @max: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-review-cross-file-v2 | 1 | 68/99 | 37/48 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: FFFFF; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: F; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: F; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: PPP; codex openai/gpt-6.1-sol: PFFPP; codex openai/gpt-6.1-sol @high: FFFFFPF; codex openai/gpt-6.1-sol @medium: PPPPPPFPPP; codex openai/gpt-6.1-sol @xhigh: FFFPF; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: FFF; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PFP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPP; pi zai/glm-5.2: P |
| hard-review-cross-file-v2 | 2 | 205/248 | 136/158 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: P; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: P; antigravity google/gemini-4-argon @low: P; antigravity google/gemini-4-argon @medium: P; claude-code anthropic/claude-haiku-5-5: P; claude-code anthropic/claude-haiku-5-5 @high: P; claude-code anthropic/claude-haiku-5-5 @low: P; claude-code anthropic/claude-haiku-5-5 @max: P; claude-code anthropic/claude-haiku-5-5 @medium: P; claude-code anthropic/claude-haiku-5-5 @xhigh: P; claude-code anthropic/claude-opus-5-5: PPPPPP; claude-code anthropic/claude-opus-5-5 @high: P; claude-code anthropic/claude-opus-5-5 @low: P; claude-code anthropic/claude-opus-5-5 @max: P; claude-code anthropic/claude-opus-5-5 @medium: P; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: PP; claude-code anthropic/claude-sonnet-5-5 @high: P; claude-code anthropic/claude-sonnet-5-5 @low: P; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: P; claude-code anthropic/claude-sonnet-5-5 @xhigh: P; claude-code baseten/glm-5.3-flash: PP; claude-code fireworks/kimi-k3: PP; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: PP; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: PP; claude-code openai/gpt-5.6-terra @high: P; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: P; claude-code openai/gpt-6-luna @max: P; claude-code openai/gpt-6-luna @xhigh: P; claude-code openai/gpt-6.1-sol: FF; claude-code openai/gpt-6.1-sol @high: F; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: F; claude-code together/deepseek-v4.1-flash: PPPP; claude-code zai/glm-5.2: PP; codex anthropic/claude-haiku-5-5 @high: P; codex anthropic/claude-haiku-5-5 @low: P; codex anthropic/claude-haiku-5-5 @max: F; codex anthropic/claude-haiku-5-5 @medium: P; codex anthropic/claude-haiku-5-5 @xhigh: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-opus-5-5 @high: P; codex anthropic/claude-opus-5-5 @low: F; codex anthropic/claude-opus-5-5 @medium: P; codex anthropic/claude-opus-5-5 @xhigh: P; codex anthropic/claude-sonnet-5-5: FP; codex anthropic/claude-sonnet-5-5 @high: P; codex anthropic/claude-sonnet-5-5 @low: P; codex anthropic/claude-sonnet-5-5 @medium: P; codex anthropic/claude-sonnet-5-5 @xhigh: P; codex baseten/glm-5.3-flash: PP; codex fireworks/kimi-k3: PPPP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: PF; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: F; codex google/gemini-3.8-flash @medium: F; codex moonshot/kimi-k3: PP; codex together/deepseek-v4.1-flash: PP; codex zai/glm-5.2: PPPP; codex openai/gpt-5.6-terra @high: P; codex openai/gpt-5.6-terra @low: P; codex openai/gpt-5.6-terra @max: P; codex openai/gpt-5.6-terra @medium: P; codex openai/gpt-5.6-terra @ultra: F; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6-luna @high: P; codex openai/gpt-6-luna @low: P; codex openai/gpt-6-luna @max: P; codex openai/gpt-6-luna @medium: P; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: PFFPP; codex openai/gpt-6.1-sol @high: FFFPPFPFPF; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: F; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @ultra: F; codex openai/gpt-6.1-sol @xhigh: FFFPF; opencode anthropic/claude-haiku-5-5 @high: P; opencode anthropic/claude-haiku-5-5 @low: P; opencode anthropic/claude-haiku-5-5 @max: F; opencode anthropic/claude-haiku-5-5 @medium: P; opencode anthropic/claude-haiku-5-5 @xhigh: F; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-opus-5-5 @high: P; opencode anthropic/claude-opus-5-5 @low: P; opencode anthropic/claude-opus-5-5 @medium: P; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: PP; opencode anthropic/claude-sonnet-5-5 @high: P; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: P; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: PPPP; opencode fireworks/kimi-k3: PP; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: PP; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: PP; opencode openai/gpt-5.6-terra @high: P; opencode openai/gpt-5.6-terra @low: P; opencode openai/gpt-6-luna @high: P; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: P; opencode openai/gpt-6-luna @xhigh: P; opencode openai/gpt-6.1-sol: FFFF; opencode openai/gpt-6.1-sol @high: F; opencode openai/gpt-6.1-sol @low: P; opencode openai/gpt-6.1-sol @max: F; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: PP; pi anthropic/claude-haiku-5-5 @high: P; pi anthropic/claude-haiku-5-5 @low: P; pi anthropic/claude-haiku-5-5 @max: F; pi anthropic/claude-haiku-5-5 @medium: P; pi anthropic/claude-haiku-5-5 @xhigh: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-opus-5-5 @high: P; pi anthropic/claude-opus-5-5 @low: P; pi anthropic/claude-opus-5-5 @medium: P; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: PP; pi anthropic/claude-sonnet-5-5 @high: P; pi anthropic/claude-sonnet-5-5 @low: P; pi anthropic/claude-sonnet-5-5 @medium: P; pi anthropic/claude-sonnet-5-5 @xhigh: P; pi baseten/glm-5.3-flash: PPPP; pi fireworks/kimi-k3: PP; pi fireworks/qwen3.8-max: PFP; pi google/gemini-3.8-flash: PP; pi google/gemini-3.8-flash @high: PF; pi google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @medium: P; pi moonshot/kimi-k3: PP; pi openai/gpt-5.6-terra @high: P; pi openai/gpt-5.6-terra @low: P; pi openai/gpt-6-luna @high: P; pi openai/gpt-6-luna @low: P; pi openai/gpt-6-luna @max: P; pi openai/gpt-6-luna @xhigh: P; pi openai/gpt-6.1-sol: FP; pi openai/gpt-6.1-sol @high: F; pi openai/gpt-6.1-sol @low: P; pi openai/gpt-6.1-sol @max: F; pi together/deepseek-v4.1-flash: PPPPPP; pi zai/glm-5.2: PP |
| hard-review-decoy-decision-replies | 1 | 36/95 | 28/47 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: FFFFF; claude-code anthropic/claude-opus-5-5 @xhigh: FFFFF; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: FFP; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: F; claude-code together/deepseek-v4.1-flash: PFP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: FFP; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: PFP; codex openai/gpt-6.1-sol: FFFFF; codex openai/gpt-6.1-sol @high: FFFFF; codex openai/gpt-6.1-sol @medium: FFFFFFFFFF; codex openai/gpt-6.1-sol @xhigh: FFFFF; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PFP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: F; opencode openai/gpt-6.1-sol: FFF; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: FPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PFP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: F; pi together/deepseek-v4.1-flash: PP; pi zai/glm-5.2: P |
| hard-review-decoy-decision-replies | 2 | 81/216 | 70/158 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: P; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: P; antigravity google/gemini-4-argon @low: F; antigravity google/gemini-4-argon @medium: P; claude-code anthropic/claude-haiku-5-5 @high: F; claude-code anthropic/claude-haiku-5-5 @low: F; claude-code anthropic/claude-haiku-5-5 @max: F; claude-code anthropic/claude-haiku-5-5 @medium: F; claude-code anthropic/claude-haiku-5-5 @xhigh: F; claude-code anthropic/claude-opus-5-5: FFFFF; claude-code anthropic/claude-opus-5-5 @high: F; claude-code anthropic/claude-opus-5-5 @low: F; claude-code anthropic/claude-opus-5-5 @max: P; claude-code anthropic/claude-opus-5-5 @medium: F; claude-code anthropic/claude-opus-5-5 @xhigh: FFFFF; claude-code anthropic/claude-sonnet-5-5: F; claude-code anthropic/claude-sonnet-5-5 @high: F; claude-code anthropic/claude-sonnet-5-5 @low: F; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: F; claude-code anthropic/claude-sonnet-5-5 @xhigh: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: FFP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-5.6-terra @high: F; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: P; claude-code openai/gpt-6-luna @max: F; claude-code openai/gpt-6-luna @xhigh: F; claude-code openai/gpt-6.1-sol: F; claude-code openai/gpt-6.1-sol @high: F; claude-code openai/gpt-6.1-sol @low: F; claude-code openai/gpt-6.1-sol @max: F; claude-code together/deepseek-v4.1-flash: PFP; claude-code zai/glm-5.2: P; codex anthropic/claude-haiku-5-5 @high: F; codex anthropic/claude-haiku-5-5 @low: P; codex anthropic/claude-haiku-5-5 @max: F; codex anthropic/claude-haiku-5-5 @medium: P; codex anthropic/claude-haiku-5-5 @xhigh: F; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-opus-5-5 @high: P; codex anthropic/claude-opus-5-5 @low: P; codex anthropic/claude-opus-5-5 @max: P; codex anthropic/claude-opus-5-5 @medium: P; codex anthropic/claude-opus-5-5 @xhigh: F; codex anthropic/claude-sonnet-5-5: P; codex anthropic/claude-sonnet-5-5 @high: F; codex anthropic/claude-sonnet-5-5 @low: F; codex anthropic/claude-sonnet-5-5 @medium: F; codex anthropic/claude-sonnet-5-5 @xhigh: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: FFP; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex google/gemini-3.8-flash @medium: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: PFP; codex openai/gpt-5.6-terra @high: F; codex openai/gpt-5.6-terra @low: P; codex openai/gpt-5.6-terra @max: F; codex openai/gpt-5.6-terra @medium: F; codex openai/gpt-5.6-terra @ultra: F; codex openai/gpt-5.6-terra @xhigh: F; codex openai/gpt-6-luna @high: P; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: F; codex openai/gpt-6-luna @medium: F; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: FFFFF; codex openai/gpt-6.1-sol @high: FFFFFFFFFF; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: F; codex openai/gpt-6.1-sol @medium: FFFFFFFFFFF; codex openai/gpt-6.1-sol @ultra: F; codex openai/gpt-6.1-sol @xhigh: FFFFF; opencode anthropic/claude-haiku-5-5 @high: F; opencode anthropic/claude-haiku-5-5 @low: F; opencode anthropic/claude-haiku-5-5 @max: F; opencode anthropic/claude-haiku-5-5 @medium: P; opencode anthropic/claude-haiku-5-5 @xhigh: F; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-opus-5-5 @high: F; opencode anthropic/claude-opus-5-5 @low: F; opencode anthropic/claude-opus-5-5 @medium: P; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: P; opencode anthropic/claude-sonnet-5-5 @high: F; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: F; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: PFP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: F; opencode openai/gpt-5.6-terra @high: F; opencode openai/gpt-5.6-terra @low: F; opencode openai/gpt-6-luna @high: P; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: F; opencode openai/gpt-6-luna @xhigh: F; opencode openai/gpt-6.1-sol: FFF; opencode openai/gpt-6.1-sol @high: F; opencode openai/gpt-6.1-sol @low: F; opencode openai/gpt-6.1-sol @max: F; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-haiku-5-5 @high: P; pi anthropic/claude-haiku-5-5 @low: P; pi anthropic/claude-haiku-5-5 @max: F; pi anthropic/claude-haiku-5-5 @medium: P; pi anthropic/claude-haiku-5-5 @xhigh: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-opus-5-5 @high: P; pi anthropic/claude-opus-5-5 @low: F; pi anthropic/claude-opus-5-5 @medium: F; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: F; pi anthropic/claude-sonnet-5-5 @high: F; pi anthropic/claude-sonnet-5-5 @low: F; pi anthropic/claude-sonnet-5-5 @medium: F; pi anthropic/claude-sonnet-5-5 @xhigh: P; pi baseten/glm-5.3-flash: FPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PFP; pi google/gemini-3.8-flash: PF; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @medium: P; pi moonshot/kimi-k3: P; pi openai/gpt-5.6-terra @high: F; pi openai/gpt-5.6-terra @low: P; pi openai/gpt-6-luna @high: F; pi openai/gpt-6-luna @low: F; pi openai/gpt-6-luna @max: F; pi openai/gpt-6-luna @xhigh: F; pi openai/gpt-6.1-sol: F; pi openai/gpt-6.1-sol @high: F; pi openai/gpt-6.1-sol @low: F; pi openai/gpt-6.1-sol @max: F; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: P |
| hard-review-decoy-job-leases | 1 | 80/94 | 38/47 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: FFFFF; claude-code anthropic/claude-opus-5-5 @xhigh: PPPP; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PFP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: F; codex zai/glm-5.2: PPP; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PP; pi zai/glm-5.2: P |
| hard-review-decoy-job-leases | 2 | 199/249 | 126/162 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: F; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: P; antigravity google/gemini-4-argon @low: P; antigravity google/gemini-4-argon @medium: P; claude-code anthropic/claude-haiku-5-5: P; claude-code anthropic/claude-haiku-5-5 @high: P; claude-code anthropic/claude-haiku-5-5 @low: P; claude-code anthropic/claude-haiku-5-5 @max: P; claude-code anthropic/claude-haiku-5-5 @medium: P; claude-code anthropic/claude-haiku-5-5 @xhigh: P; claude-code anthropic/claude-opus-5-5: FFFFF; claude-code anthropic/claude-opus-5-5 @high: F; claude-code anthropic/claude-opus-5-5 @low: F; claude-code anthropic/claude-opus-5-5 @max: PP; claude-code anthropic/claude-opus-5-5 @medium: F; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: P; claude-code anthropic/claude-sonnet-5-5 @high: P; claude-code anthropic/claude-sonnet-5-5 @low: F; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: F; claude-code anthropic/claude-sonnet-5-5 @xhigh: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-5.6-terra @high: P; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: P; claude-code openai/gpt-6-luna @max: F; claude-code openai/gpt-6-luna @xhigh: P; claude-code openai/gpt-6.1-sol: P; claude-code openai/gpt-6.1-sol @high: PP; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-haiku-5-5 @high: F; codex anthropic/claude-haiku-5-5 @low: P; codex anthropic/claude-haiku-5-5 @max: FP; codex anthropic/claude-haiku-5-5 @medium: FP; codex anthropic/claude-haiku-5-5 @xhigh: PP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-opus-5-5 @high: PP; codex anthropic/claude-opus-5-5 @low: PF; codex anthropic/claude-opus-5-5 @max: PP; codex anthropic/claude-opus-5-5 @medium: FF; codex anthropic/claude-opus-5-5 @xhigh: PP; codex anthropic/claude-sonnet-5-5: F; codex anthropic/claude-sonnet-5-5 @high: P; codex anthropic/claude-sonnet-5-5 @low: F; codex anthropic/claude-sonnet-5-5 @max: PP; codex anthropic/claude-sonnet-5-5 @medium: FF; codex anthropic/claude-sonnet-5-5 @xhigh: PP; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PFP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex google/gemini-3.8-flash @medium: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: PPP; codex openai/gpt-5.6-terra @high: P; codex openai/gpt-5.6-terra @low: P; codex openai/gpt-5.6-terra @max: P; codex openai/gpt-5.6-terra @medium: P; codex openai/gpt-5.6-terra @ultra: P; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6-luna @high: P; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: P; codex openai/gpt-6-luna @medium: P; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: P; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @ultra: PP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-haiku-5-5 @high: P; opencode anthropic/claude-haiku-5-5 @low: P; opencode anthropic/claude-haiku-5-5 @max: P; opencode anthropic/claude-haiku-5-5 @medium: P; opencode anthropic/claude-haiku-5-5 @xhigh: P; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-opus-5-5 @high: PP; opencode anthropic/claude-opus-5-5 @low: FF; opencode anthropic/claude-opus-5-5 @max: P; opencode anthropic/claude-opus-5-5 @medium: F; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: P; opencode anthropic/claude-sonnet-5-5 @high: F; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: F; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: F; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-5.6-terra @high: P; opencode openai/gpt-5.6-terra @low: P; opencode openai/gpt-6-luna @high: F; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: FF; opencode openai/gpt-6-luna @xhigh: PP; opencode openai/gpt-6.1-sol: PPP; opencode openai/gpt-6.1-sol @high: PP; opencode openai/gpt-6.1-sol @low: PP; opencode openai/gpt-6.1-sol @max: PP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-haiku-5-5 @high: P; pi anthropic/claude-haiku-5-5 @low: P; pi anthropic/claude-haiku-5-5 @max: F; pi anthropic/claude-haiku-5-5 @medium: P; pi anthropic/claude-haiku-5-5 @xhigh: P; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-opus-5-5 @high: PF; pi anthropic/claude-opus-5-5 @low: FF; pi anthropic/claude-opus-5-5 @max: P; pi anthropic/claude-opus-5-5 @medium: P; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: P; pi anthropic/claude-sonnet-5-5 @high: P; pi anthropic/claude-sonnet-5-5 @low: F; pi anthropic/claude-sonnet-5-5 @medium: F; pi anthropic/claude-sonnet-5-5 @xhigh: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: PPPPP; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: F; pi google/gemini-3.8-flash @medium: P; pi moonshot/kimi-k3: P; pi openai/gpt-5.6-terra @high: P; pi openai/gpt-5.6-terra @low: P; pi openai/gpt-6-luna @high: P; pi openai/gpt-6-luna @low: F; pi openai/gpt-6-luna @max: P; pi openai/gpt-6-luna @xhigh: F; pi openai/gpt-6.1-sol: P; pi openai/gpt-6.1-sol @high: PP; pi openai/gpt-6.1-sol @low: PP; pi openai/gpt-6.1-sol @max: PP; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: P |
| hard-review-multi-defect-outbox | 1 | 46/46 | 43/43 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi zai/glm-5.2: P |
| hard-review-multi-defect-outbox | 2 | 151/162 | 148/159 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: P; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: P; antigravity google/gemini-4-argon @low: P; antigravity google/gemini-4-argon @medium: P; claude-code anthropic/claude-haiku-5-5 @high: P; claude-code anthropic/claude-haiku-5-5 @low: P; claude-code anthropic/claude-haiku-5-5 @max: P; claude-code anthropic/claude-haiku-5-5 @medium: P; claude-code anthropic/claude-haiku-5-5 @xhigh: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-opus-5-5 @high: P; claude-code anthropic/claude-opus-5-5 @low: F; claude-code anthropic/claude-opus-5-5 @max: P; claude-code anthropic/claude-opus-5-5 @medium: P; claude-code anthropic/claude-opus-5-5 @xhigh: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code anthropic/claude-sonnet-5-5 @high: P; claude-code anthropic/claude-sonnet-5-5 @low: P; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: P; claude-code anthropic/claude-sonnet-5-5 @xhigh: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-5.6-terra @high: P; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: P; claude-code openai/gpt-6-luna @max: P; claude-code openai/gpt-6-luna @xhigh: F; claude-code openai/gpt-6.1-sol: P; claude-code openai/gpt-6.1-sol @high: P; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-haiku-5-5 @high: P; codex anthropic/claude-haiku-5-5 @low: P; codex anthropic/claude-haiku-5-5 @max: P; codex anthropic/claude-haiku-5-5 @medium: P; codex anthropic/claude-haiku-5-5 @xhigh: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-opus-5-5 @high: P; codex anthropic/claude-opus-5-5 @low: F; codex anthropic/claude-opus-5-5 @max: P; codex anthropic/claude-opus-5-5 @medium: P; codex anthropic/claude-opus-5-5 @xhigh: P; codex anthropic/claude-sonnet-5-5: P; codex anthropic/claude-sonnet-5-5 @high: P; codex anthropic/claude-sonnet-5-5 @low: F; codex anthropic/claude-sonnet-5-5 @max: P; codex anthropic/claude-sonnet-5-5 @medium: F; codex anthropic/claude-sonnet-5-5 @xhigh: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: P; codex google/gemini-3.8-flash @medium: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-5.6-terra @high: P; codex openai/gpt-5.6-terra @low: P; codex openai/gpt-5.6-terra @max: P; codex openai/gpt-5.6-terra @medium: P; codex openai/gpt-5.6-terra @ultra: P; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6-luna @high: P; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: P; codex openai/gpt-6-luna @medium: P; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: P; codex openai/gpt-6.1-sol @medium: P; codex openai/gpt-6.1-sol @ultra: P; codex openai/gpt-6.1-sol @xhigh: P; opencode anthropic/claude-haiku-5-5 @high: P; opencode anthropic/claude-haiku-5-5 @low: P; opencode anthropic/claude-haiku-5-5 @max: P; opencode anthropic/claude-haiku-5-5 @medium: P; opencode anthropic/claude-haiku-5-5 @xhigh: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-opus-5-5 @high: P; opencode anthropic/claude-opus-5-5 @low: F; opencode anthropic/claude-opus-5-5 @medium: P; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: P; opencode anthropic/claude-sonnet-5-5 @high: P; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: F; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-5.6-terra @high: P; opencode openai/gpt-5.6-terra @low: P; opencode openai/gpt-6-luna @high: P; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: P; opencode openai/gpt-6-luna @xhigh: P; opencode openai/gpt-6.1-sol: P; opencode openai/gpt-6.1-sol @high: P; opencode openai/gpt-6.1-sol @low: P; opencode openai/gpt-6.1-sol @max: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-haiku-5-5 @high: P; pi anthropic/claude-haiku-5-5 @low: P; pi anthropic/claude-haiku-5-5 @max: P; pi anthropic/claude-haiku-5-5 @medium: P; pi anthropic/claude-haiku-5-5 @xhigh: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-opus-5-5 @high: P; pi anthropic/claude-opus-5-5 @low: F; pi anthropic/claude-opus-5-5 @medium: P; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: P; pi anthropic/claude-sonnet-5-5 @high: P; pi anthropic/claude-sonnet-5-5 @low: F; pi anthropic/claude-sonnet-5-5 @medium: P; pi anthropic/claude-sonnet-5-5 @xhigh: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi google/gemini-3.8-flash @medium: P; pi moonshot/kimi-k3: P; pi openai/gpt-5.6-terra @high: P; pi openai/gpt-5.6-terra @low: P; pi openai/gpt-6-luna @high: P; pi openai/gpt-6-luna @low: P; pi openai/gpt-6-luna @max: P; pi openai/gpt-6-luna @xhigh: P; pi openai/gpt-6.1-sol: P; pi openai/gpt-6.1-sol @high: P; pi openai/gpt-6.1-sol @low: P; pi openai/gpt-6.1-sol @max: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| hard-review-multi-defect-outbox-v2 | 1 | 71/93 | 39/48 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PFPFP; claude-code anthropic/claude-opus-5-5 @xhigh: P; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PFP; claude-code google/gemini-3.8-flash: F; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PPF; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: PPP; codex openai/gpt-6.1-sol: FPPPP; codex openai/gpt-6.1-sol @high: PFPPF; codex openai/gpt-6.1-sol @medium: PPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: FPPPF; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PPF; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: FFP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPP; pi zai/glm-5.2: P |
| hard-review-multi-defect-outbox-v2 | 2 | 136/218 | 88/157 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: F; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: P; antigravity google/gemini-4-argon @low: P; antigravity google/gemini-4-argon @medium: P; claude-code anthropic/claude-haiku-5-5 @high: P; claude-code anthropic/claude-haiku-5-5 @low: P; claude-code anthropic/claude-haiku-5-5 @max: F; claude-code anthropic/claude-haiku-5-5 @medium: P; claude-code anthropic/claude-haiku-5-5 @xhigh: P; claude-code anthropic/claude-opus-5-5: PFPFP; claude-code anthropic/claude-opus-5-5 @high: F; claude-code anthropic/claude-opus-5-5 @low: F; claude-code anthropic/claude-opus-5-5 @max: P; claude-code anthropic/claude-opus-5-5 @medium: F; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: P; claude-code anthropic/claude-sonnet-5-5 @high: F; claude-code anthropic/claude-sonnet-5-5 @low: F; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: F; claude-code anthropic/claude-sonnet-5-5 @xhigh: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PFP; claude-code google/gemini-3.8-flash: F; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-5.6-terra @high: P; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: F; claude-code openai/gpt-6-luna @max: F; claude-code openai/gpt-6-luna @xhigh: F; claude-code openai/gpt-6.1-sol: P; claude-code openai/gpt-6.1-sol @high: F; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: F; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-haiku-5-5 @high: F; codex anthropic/claude-haiku-5-5 @low: F; codex anthropic/claude-haiku-5-5 @max: F; codex anthropic/claude-haiku-5-5 @medium: F; codex anthropic/claude-haiku-5-5 @xhigh: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-opus-5-5 @high: P; codex anthropic/claude-opus-5-5 @low: F; codex anthropic/claude-opus-5-5 @medium: P; codex anthropic/claude-opus-5-5 @xhigh: P; codex anthropic/claude-sonnet-5-5: P; codex anthropic/claude-sonnet-5-5 @high: P; codex anthropic/claude-sonnet-5-5 @low: F; codex anthropic/claude-sonnet-5-5 @medium: F; codex anthropic/claude-sonnet-5-5 @xhigh: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: PPF; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: P; codex google/gemini-3.8-flash @low: F; codex google/gemini-3.8-flash @medium: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: PPP; codex openai/gpt-5.6-terra @high: P; codex openai/gpt-5.6-terra @low: F; codex openai/gpt-5.6-terra @max: F; codex openai/gpt-5.6-terra @medium: P; codex openai/gpt-5.6-terra @ultra: F; codex openai/gpt-5.6-terra @xhigh: F; codex openai/gpt-6-luna @high: P; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: F; codex openai/gpt-6-luna @medium: F; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: FPPPP; codex openai/gpt-6.1-sol @high: FPPPPPPPPF; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: F; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @ultra: F; codex openai/gpt-6.1-sol @xhigh: FPPPF; opencode anthropic/claude-haiku-5-5 @high: F; opencode anthropic/claude-haiku-5-5 @low: F; opencode anthropic/claude-haiku-5-5 @max: F; opencode anthropic/claude-haiku-5-5 @medium: F; opencode anthropic/claude-haiku-5-5 @xhigh: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-opus-5-5 @high: F; opencode anthropic/claude-opus-5-5 @low: F; opencode anthropic/claude-opus-5-5 @medium: F; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: P; opencode anthropic/claude-sonnet-5-5 @high: P; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: F; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: PPF; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: F; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-5.6-terra @high: F; opencode openai/gpt-5.6-terra @low: F; opencode openai/gpt-6-luna @high: P; opencode openai/gpt-6-luna @low: F; opencode openai/gpt-6-luna @max: F; opencode openai/gpt-6-luna @xhigh: F; opencode openai/gpt-6.1-sol: FFP; opencode openai/gpt-6.1-sol @high: F; opencode openai/gpt-6.1-sol @low: P; opencode openai/gpt-6.1-sol @max: F; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-haiku-5-5 @high: P; pi anthropic/claude-haiku-5-5 @low: F; pi anthropic/claude-haiku-5-5 @max: P; pi anthropic/claude-haiku-5-5 @medium: F; pi anthropic/claude-haiku-5-5 @xhigh: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-opus-5-5 @high: F; pi anthropic/claude-opus-5-5 @low: F; pi anthropic/claude-opus-5-5 @medium: F; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: P; pi anthropic/claude-sonnet-5-5 @high: P; pi anthropic/claude-sonnet-5-5 @low: F; pi anthropic/claude-sonnet-5-5 @medium: P; pi anthropic/claude-sonnet-5-5 @xhigh: P; pi baseten/glm-5.3-flash: FPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: PPPPP; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: F; pi google/gemini-3.8-flash @medium: F; pi moonshot/kimi-k3: P; pi openai/gpt-5.6-terra @high: P; pi openai/gpt-5.6-terra @low: F; pi openai/gpt-6-luna @high: F; pi openai/gpt-6-luna @low: F; pi openai/gpt-6-luna @max: F; pi openai/gpt-6-luna @xhigh: F; pi openai/gpt-6.1-sol: P; pi openai/gpt-6.1-sol @high: F; pi openai/gpt-6.1-sol @low: P; pi openai/gpt-6.1-sol @max: F; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: P |
| hard-review-multi-defect-tool-calls | 1 | 5/5 | 5/5 | no | antigravity google/gemini-3.1-pro @high: P; claude-code fireworks/qwen3.8-max: P; claude-code zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P |
| hard-review-multi-defect-tool-calls | 2 | 77/91 | 36/47 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: F; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPF; claude-code google/gemini-3.8-flash: F; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: PPP; codex openai/gpt-6.1-sol: PPPP; codex openai/gpt-6.1-sol @high: PPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: FPP; pi google/gemini-3.8-flash: F; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PFP; pi zai/glm-5.2: F |
| hard-review-multi-defect-tool-calls | 3 | 177/221 | 117/156 | yes | antigravity google/gemini-3.1-pro @high: PP; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-3.8-flash @low: F; antigravity google/gemini-3.8-flash @medium: P; antigravity google/gemini-4-argon @high: F; antigravity google/gemini-4-argon @low: P; antigravity google/gemini-4-argon @medium: F; claude-code anthropic/claude-haiku-5-5 @high: F; claude-code anthropic/claude-haiku-5-5 @low: P; claude-code anthropic/claude-haiku-5-5 @max: F; claude-code anthropic/claude-haiku-5-5 @medium: P; claude-code anthropic/claude-haiku-5-5 @xhigh: F; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @high: P; claude-code anthropic/claude-opus-5-5 @low: F; claude-code anthropic/claude-opus-5-5 @max: P; claude-code anthropic/claude-opus-5-5 @medium: P; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPF; claude-code anthropic/claude-sonnet-5-5: P; claude-code anthropic/claude-sonnet-5-5 @high: P; claude-code anthropic/claude-sonnet-5-5 @low: P; claude-code anthropic/claude-sonnet-5-5 @max: P; claude-code anthropic/claude-sonnet-5-5 @medium: P; claude-code anthropic/claude-sonnet-5-5 @xhigh: F; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPPF; claude-code google/gemini-3.8-flash: F; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-5.6-terra @high: P; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: P; claude-code openai/gpt-6-luna @low: F; claude-code openai/gpt-6-luna @max: P; claude-code openai/gpt-6-luna @xhigh: F; claude-code openai/gpt-6.1-sol: P; claude-code openai/gpt-6.1-sol @high: P; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: PP; codex anthropic/claude-haiku-5-5 @high: P; codex anthropic/claude-haiku-5-5 @low: P; codex anthropic/claude-haiku-5-5 @max: F; codex anthropic/claude-haiku-5-5 @medium: P; codex anthropic/claude-haiku-5-5 @xhigh: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-opus-5-5 @high: P; codex anthropic/claude-opus-5-5 @low: P; codex anthropic/claude-opus-5-5 @medium: P; codex anthropic/claude-opus-5-5 @xhigh: P; codex anthropic/claude-sonnet-5-5: F; codex anthropic/claude-sonnet-5-5 @high: P; codex anthropic/claude-sonnet-5-5 @low: F; codex anthropic/claude-sonnet-5-5 @medium: P; codex anthropic/claude-sonnet-5-5 @xhigh: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @high: F; codex google/gemini-3.8-flash @low: F; codex google/gemini-3.8-flash @medium: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: PP; codex zai/glm-5.2: PPP; codex openai/gpt-5.6-terra @high: F; codex openai/gpt-5.6-terra @low: P; codex openai/gpt-5.6-terra @max: F; codex openai/gpt-5.6-terra @medium: F; codex openai/gpt-5.6-terra @ultra: F; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6-luna @high: P; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: P; codex openai/gpt-6-luna @medium: P; codex openai/gpt-6-luna @xhigh: P; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @low: P; codex openai/gpt-6.1-sol @max: P; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @ultra: P; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-haiku-5-5 @high: P; opencode anthropic/claude-haiku-5-5 @low: P; opencode anthropic/claude-haiku-5-5 @max: F; opencode anthropic/claude-haiku-5-5 @medium: F; opencode anthropic/claude-haiku-5-5 @xhigh: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-opus-5-5 @high: P; opencode anthropic/claude-opus-5-5 @low: F; opencode anthropic/claude-opus-5-5 @medium: P; opencode anthropic/claude-opus-5-5 @xhigh: P; opencode anthropic/claude-sonnet-5-5: P; opencode anthropic/claude-sonnet-5-5 @high: P; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: F; opencode anthropic/claude-sonnet-5-5 @xhigh: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode google/gemini-3.8-flash @medium: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-5.6-terra @high: P; opencode openai/gpt-5.6-terra @low: P; opencode openai/gpt-6-luna @high: P; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: F; opencode openai/gpt-6-luna @xhigh: P; opencode openai/gpt-6.1-sol: PPP; opencode openai/gpt-6.1-sol @high: P; opencode openai/gpt-6.1-sol @low: P; opencode openai/gpt-6.1-sol @max: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-haiku-5-5 @high: P; pi anthropic/claude-haiku-5-5 @low: P; pi anthropic/claude-haiku-5-5 @max: F; pi anthropic/claude-haiku-5-5 @medium: F; pi anthropic/claude-haiku-5-5 @xhigh: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-opus-5-5 @high: P; pi anthropic/claude-opus-5-5 @low: P; pi anthropic/claude-opus-5-5 @medium: P; pi anthropic/claude-opus-5-5 @xhigh: P; pi anthropic/claude-sonnet-5-5: F; pi anthropic/claude-sonnet-5-5 @high: P; pi anthropic/claude-sonnet-5-5 @low: F; pi anthropic/claude-sonnet-5-5 @medium: P; pi anthropic/claude-sonnet-5-5 @xhigh: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: FPP; pi google/gemini-3.8-flash: PFPP; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @medium: P; pi moonshot/kimi-k3: P; pi openai/gpt-5.6-terra @high: F; pi openai/gpt-5.6-terra @low: P; pi openai/gpt-6-luna @high: P; pi openai/gpt-6-luna @low: F; pi openai/gpt-6-luna @max: P; pi openai/gpt-6-luna @xhigh: P; pi openai/gpt-6.1-sol: P; pi openai/gpt-6.1-sol @high: P; pi openai/gpt-6.1-sol @low: P; pi openai/gpt-6.1-sol @max: P; pi together/deepseek-v4.1-flash: PPFPP; pi zai/glm-5.2: F |
| hard-review-multi-defect-tool-calls-v2 | 1 | 6/47 | 6/45 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: FFF; antigravity google/gemini-4-argon @high: F; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: F; claude-code google/gemini-3.8-flash: F; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: F; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: F; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: F; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: F; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: F; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: F; opencode moonshot/kimi-k3: F; opencode openai/gpt-6.1-sol: F; opencode together/deepseek-v4.1-flash: F; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: F; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: F; pi google/gemini-3.8-flash @high: F; pi moonshot/kimi-k3: F; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: F |
| hard-review-multi-defect-tool-calls-v2 | 2 | 29/159 | 29/157 | yes | antigravity google/gemini-3.1-pro @high: F; antigravity google/gemini-3.8-flash @high: FFF; antigravity google/gemini-3.8-flash @low: F; antigravity google/gemini-3.8-flash @medium: F; antigravity google/gemini-4-argon @high: F; antigravity google/gemini-4-argon @low: F; antigravity google/gemini-4-argon @medium: F; claude-code anthropic/claude-haiku-5-5 @high: F; claude-code anthropic/claude-haiku-5-5 @low: F; claude-code anthropic/claude-haiku-5-5 @max: F; claude-code anthropic/claude-haiku-5-5 @medium: F; claude-code anthropic/claude-haiku-5-5 @xhigh: F; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-opus-5-5 @high: P; claude-code anthropic/claude-opus-5-5 @low: F; claude-code anthropic/claude-opus-5-5 @max: F; claude-code anthropic/claude-opus-5-5 @medium: P; claude-code anthropic/claude-opus-5-5 @xhigh: F; claude-code anthropic/claude-sonnet-5-5: F; claude-code anthropic/claude-sonnet-5-5 @high: F; claude-code anthropic/claude-sonnet-5-5 @low: F; claude-code anthropic/claude-sonnet-5-5 @max: F; claude-code anthropic/claude-sonnet-5-5 @medium: F; claude-code anthropic/claude-sonnet-5-5 @xhigh: P; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: F; claude-code google/gemini-3.8-flash: F; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: F; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-5.6-terra @high: F; claude-code openai/gpt-5.6-terra @low: P; claude-code openai/gpt-6-luna @high: F; claude-code openai/gpt-6-luna @low: F; claude-code openai/gpt-6-luna @max: F; claude-code openai/gpt-6-luna @xhigh: P; claude-code openai/gpt-6.1-sol: P; claude-code openai/gpt-6.1-sol @high: P; claude-code openai/gpt-6.1-sol @low: P; claude-code openai/gpt-6.1-sol @max: F; claude-code together/deepseek-v4.1-flash: F; claude-code zai/glm-5.2: F; codex anthropic/claude-haiku-5-5 @high: F; codex anthropic/claude-haiku-5-5 @low: F; codex anthropic/claude-haiku-5-5 @max: F; codex anthropic/claude-haiku-5-5 @medium: F; codex anthropic/claude-haiku-5-5 @xhigh: F; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-opus-5-5 @high: P; codex anthropic/claude-opus-5-5 @low: F; codex anthropic/claude-opus-5-5 @max: F; codex anthropic/claude-opus-5-5 @medium: P; codex anthropic/claude-opus-5-5 @xhigh: F; codex anthropic/claude-sonnet-5-5: F; codex anthropic/claude-sonnet-5-5 @high: F; codex anthropic/claude-sonnet-5-5 @low: F; codex anthropic/claude-sonnet-5-5 @medium: F; codex anthropic/claude-sonnet-5-5 @xhigh: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: F; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: F; codex google/gemini-3.8-flash @high: F; codex google/gemini-3.8-flash @low: F; codex google/gemini-3.8-flash @medium: F; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: F; codex openai/gpt-5.6-terra @high: P; codex openai/gpt-5.6-terra @low: F; codex openai/gpt-5.6-terra @max: F; codex openai/gpt-5.6-terra @medium: P; codex openai/gpt-5.6-terra @ultra: F; codex openai/gpt-5.6-terra @xhigh: P; codex openai/gpt-6-luna @high: F; codex openai/gpt-6-luna @low: F; codex openai/gpt-6-luna @max: P; codex openai/gpt-6-luna @medium: F; codex openai/gpt-6-luna @xhigh: F; codex openai/gpt-6.1-sol: F; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @low: F; codex openai/gpt-6.1-sol @max: P; codex openai/gpt-6.1-sol @medium: F; codex openai/gpt-6.1-sol @ultra: F; codex openai/gpt-6.1-sol @xhigh: F; opencode anthropic/claude-haiku-5-5 @high: F; opencode anthropic/claude-haiku-5-5 @low: F; opencode anthropic/claude-haiku-5-5 @max: F; opencode anthropic/claude-haiku-5-5 @medium: F; opencode anthropic/claude-haiku-5-5 @xhigh: F; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-opus-5-5 @high: F; opencode anthropic/claude-opus-5-5 @low: P; opencode anthropic/claude-opus-5-5 @medium: F; opencode anthropic/claude-opus-5-5 @xhigh: F; opencode anthropic/claude-sonnet-5-5: F; opencode anthropic/claude-sonnet-5-5 @high: F; opencode anthropic/claude-sonnet-5-5 @low: F; opencode anthropic/claude-sonnet-5-5 @medium: F; opencode anthropic/claude-sonnet-5-5 @xhigh: F; opencode baseten/glm-5.3-flash: F; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: F; opencode google/gemini-3.8-flash @high: F; opencode google/gemini-3.8-flash @low: F; opencode google/gemini-3.8-flash @medium: F; opencode moonshot/kimi-k3: F; opencode openai/gpt-5.6-terra @high: P; opencode openai/gpt-5.6-terra @low: P; opencode openai/gpt-6-luna @high: F; opencode openai/gpt-6-luna @low: P; opencode openai/gpt-6-luna @max: F; opencode openai/gpt-6-luna @xhigh: P; opencode openai/gpt-6.1-sol: F; opencode openai/gpt-6.1-sol @high: F; opencode openai/gpt-6.1-sol @low: F; opencode openai/gpt-6.1-sol @max: F; opencode together/deepseek-v4.1-flash: F; opencode zai/glm-5.2: F; pi anthropic/claude-haiku-5-5 @high: F; pi anthropic/claude-haiku-5-5 @low: F; pi anthropic/claude-haiku-5-5 @max: F; pi anthropic/claude-haiku-5-5 @medium: F; pi anthropic/claude-haiku-5-5 @xhigh: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-opus-5-5 @high: F; pi anthropic/claude-opus-5-5 @low: F; pi anthropic/claude-opus-5-5 @medium: P; pi anthropic/claude-opus-5-5 @xhigh: F; pi anthropic/claude-sonnet-5-5: F; pi anthropic/claude-sonnet-5-5 @high: P; pi anthropic/claude-sonnet-5-5 @low: F; pi anthropic/claude-sonnet-5-5 @medium: F; pi anthropic/claude-sonnet-5-5 @xhigh: F; pi baseten/glm-5.3-flash: F; pi fireworks/kimi-k3: F; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: F; pi google/gemini-3.8-flash @high: F; pi google/gemini-3.8-flash @medium: F; pi moonshot/kimi-k3: F; pi openai/gpt-5.6-terra @high: F; pi openai/gpt-5.6-terra @low: F; pi openai/gpt-6-luna @high: P; pi openai/gpt-6-luna @low: P; pi openai/gpt-6-luna @max: F; pi openai/gpt-6-luna @xhigh: F; pi openai/gpt-6.1-sol: P; pi openai/gpt-6.1-sol @high: F; pi openai/gpt-6.1-sol @low: F; pi openai/gpt-6.1-sol @max: F; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: F |
rca
| task | version | passed / n | combos passing | discriminates | runs (P passed, F not passed) |
|---|---|---|---|---|---|
| rca-admit-pressure-crash-loop | 1 | 61/67 | 55/55 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: FFP; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: PPF; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: PF; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-5.6-luna @medium: P; codex openai/gpt-5.6-sol @high: P; codex openai/gpt-6.1-sol: FFP; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: PP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: PP; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| rca-fk-drain | 1 | 111/121 | 54/57 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: F; claude-code anthropic/claude-opus-5-5: FPPPF; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5: F; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: PPF; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: FF; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: PPP; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PPP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: PF; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: PPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: FPF; pi google/gemini-3.8-flash: PPPPP; pi google/gemini-3.8-flash @high: P; pi google/gemini-3.8-flash @low: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: P |
| rca-litellm-watchdog-restart-storm | 1 | 66/68 | 54/56 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: P; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-haiku-4-5: PP; claude-code anthropic/claude-opus-5-5: PPP; claude-code anthropic/claude-sonnet-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: PPP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: PP; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-5.6-luna @xhigh: P; codex openai/gpt-6-luna @high: F; codex openai/gpt-6.1-sol: PPP; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: PP; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
review
| task | version | passed / n | combos passing | discriminates | runs (P passed, F not passed) |
|---|---|---|---|---|---|
| review-clean-retention-plan | 1 | 46/47 | 43/44 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi zai/glm-5.2: P |
| review-clean-retention-plan | 2 | 56/56 | 53/53 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| review-corrupt-refresh-checkpoint | 1 | 8/12 | 8/12 | yes | antigravity google/gemini-3.1-pro @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code fireworks/qwen3.8-27b: F; claude-code fireworks/qwen3.8-max: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: F; codex zai/glm-5.2: F; codex openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: F; pi zai/glm-5.2: P |
| review-corrupt-refresh-checkpoint | 2 | 79/88 | 41/47 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPP; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-max: PPP; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PP; claude-code zai/glm-5.2: F; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: FP; codex openai/gpt-6.1-sol: PPPP; codex openai/gpt-6.1-sol @high: PPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PFP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPP; pi zai/glm-5.2: F |
| review-corrupt-refresh-checkpoint | 3 | 108/124 | 47/55 | yes | antigravity google/gemini-3.1-pro @high: PP; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: PP; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: F; claude-code fireworks/qwen3.8-27b: F; claude-code fireworks/qwen3.8-max: PPPP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPP; claude-code zai/glm-5.2: FF; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: PPP; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: FP; codex zai/glm-5.2: FFP; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: PFP; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: F; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: FF; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FPP; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PPP; pi google/gemini-3.8-flash: PPPPP; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPPPP; pi zai/glm-5.2: PF |
| review-publish-guard-order | 1 | 47/47 | 44/44 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi zai/glm-5.2: P |
| review-publish-guard-order | 2 | 56/56 | 53/53 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| review-rescue-local-tip | 1 | 6/10 | 6/10 | yes | antigravity google/gemini-3.1-pro @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code fireworks/qwen3.8-max: F; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: F; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; opencode zai/glm-5.2: F; pi zai/glm-5.2: F |
| review-rescue-local-tip | 2 | 56/88 | 27/47 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPP; claude-code anthropic/claude-sonnet-5-5: F; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: FFF; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PF; claude-code zai/glm-5.2: F; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PFF; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: FF; codex openai/gpt-6.1-sol: PPPP; codex openai/gpt-6.1-sol @high: PPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: FFP; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: F; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: F; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FFF; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PFF; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPP; pi zai/glm-5.2: F |
| review-rescue-local-tip | 3 | 82/121 | 34/54 | yes | antigravity google/gemini-3.1-pro @high: PP; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PPPPP; claude-code anthropic/claude-opus-5-5 @xhigh: PPPPP; claude-code anthropic/claude-sonnet-5-5: PP; claude-code baseten/glm-5.3-flash: F; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: FFFF; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: F; claude-code google/gemini-3.8-flash @low: F; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: F; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PPF; claude-code zai/glm-5.2: FF; codex anthropic/claude-opus-5-5: F; codex anthropic/claude-sonnet-5-5: F; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: PFF; codex fireworks/qwen3.8-max: F; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: F; codex together/deepseek-v4.1-flash: FF; codex zai/glm-5.2: FPF; codex openai/gpt-6.1-sol: PPPPP; codex openai/gpt-6.1-sol @high: PPPPPPPPPP; codex openai/gpt-6.1-sol @medium: PPPPPPPPPPP; codex openai/gpt-6.1-sol @xhigh: PPPPP; opencode anthropic/claude-opus-5-5: F; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: FFP; opencode fireworks/kimi-k3: F; opencode fireworks/qwen3.8-max: F; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: F; opencode openai/gpt-6.1-sol: PPP; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: FF; pi anthropic/claude-opus-5-5: F; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: FFF; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: PFF; pi google/gemini-3.8-flash: PPPP; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PPFPP; pi zai/glm-5.2: FF |
| review-retry-empty-page | 1 | 47/47 | 44/44 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi zai/glm-5.2: P |
| review-retry-empty-page | 2 | 56/56 | 53/53 | no | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| review-stale-lock-recovery | 1 | 10/12 | 10/12 | yes | antigravity google/gemini-3.1-pro @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: F; pi zai/glm-5.2: F |
| review-stale-lock-recovery | 2 | 45/47 | 43/45 | yes | antigravity google/gemini-3.1-pro @high: P; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: P; claude-code anthropic/claude-sonnet-5-5: P; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-max: P; claude-code google/gemini-3.8-flash: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: P; claude-code zai/glm-5.2: P; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: F; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: F; codex zai/glm-5.2: P; codex openai/gpt-6.1-sol: P; codex openai/gpt-6.1-sol @high: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: P; opencode zai/glm-5.2: P; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: P; pi zai/glm-5.2: P |
| review-stale-lock-recovery | 3 | 65/68 | 53/53 | yes | antigravity google/gemini-3.1-pro @high: PP; antigravity google/gemini-3.8-flash @high: PPP; antigravity google/gemini-4-argon @high: P; claude-code anthropic/claude-opus-5-5: PP; claude-code anthropic/claude-sonnet-5-5: PP; claude-code baseten/glm-5.3-flash: P; claude-code fireworks/kimi-k3: P; claude-code fireworks/qwen3.8-27b: P; claude-code fireworks/qwen3.8-max: PP; claude-code google/gemini-3.8-flash: P; claude-code google/gemini-3.8-flash @high: P; claude-code google/gemini-3.8-flash @low: P; claude-code google/gemini-3.8-flash @medium: P; claude-code moonshot/kimi-k3: P; claude-code openai/gpt-6.1-sol: P; claude-code together/deepseek-v4.1-flash: PP; claude-code zai/glm-5.2: PP; codex anthropic/claude-opus-5-5: P; codex anthropic/claude-sonnet-5-5: P; codex baseten/glm-5.3-flash: P; codex fireworks/kimi-k3: P; codex fireworks/qwen3.8-max: P; codex google/gemini-3.8-flash: P; codex google/gemini-3.8-flash @low: P; codex moonshot/kimi-k3: P; codex together/deepseek-v4.1-flash: FP; codex zai/glm-5.2: PP; codex openai/gpt-6.1-sol: PP; codex openai/gpt-6.1-sol @high: P; codex openai/gpt-6.1-sol @medium: P; opencode anthropic/claude-opus-5-5: P; opencode anthropic/claude-sonnet-5-5: P; opencode baseten/glm-5.3-flash: P; opencode fireworks/kimi-k3: P; opencode fireworks/qwen3.8-max: P; opencode google/gemini-3.8-flash: P; opencode google/gemini-3.8-flash @high: P; opencode google/gemini-3.8-flash @low: P; opencode moonshot/kimi-k3: P; opencode openai/gpt-6.1-sol: P; opencode together/deepseek-v4.1-flash: PP; opencode zai/glm-5.2: PF; pi anthropic/claude-opus-5-5: P; pi anthropic/claude-sonnet-5-5: P; pi baseten/glm-5.3-flash: P; pi fireworks/kimi-k3: P; pi fireworks/qwen3.8-max: P; pi google/gemini-3.8-flash: P; pi google/gemini-3.8-flash @high: P; pi moonshot/kimi-k3: P; pi openai/gpt-6.1-sol: P; pi together/deepseek-v4.1-flash: PP; pi zai/glm-5.2: PF |