Board · task version 1

docs

Each board ranks harness × model combinations on one suite by pass rate with a 95% Wilson interval; rows are not comparable across suites or task versions.

6 tasks
  • docs-cache-config
  • docs-config-loader-keys
  • docs-gym-baseline-keys
  • docs-gym-run-flags
  • docs-retry-behavior-diff
  • docs-snapshot-new-command
docs · task version 1
#Liverymodel · harness · configpassed / n95% Wilsonmeantaskstokens / passcost / passexcluded
1google/gemini-3.8-flashpi · config da41c66e4 / 6 30%–90%0.946262k$0.17 list0
2google/gemini-3.8-flash @highantigravity · config 034ed3a13 / 6 19%–81%0.946726k$0.37 list0
3anthropic/claude-opus-5-5opencode · config 357da8913 / 6 19%–81%0.916107k$0.22 list1
4zai/glm-5.2pi · config da41c66e3 / 6 19%–81%0.91674k$0.09 list0
5fireworks/qwen3.8-maxclaude-code · config ecb67e4e7 / 18 20%–61%0.936not measured (0/18)unknown12
6anthropic/claude-sonnet-5-5claude-code · config c32a950b4 / 12 14%–61%0.926not measured (0/12)unknown6
7fireworks/kimi-k3claude-code · config ecb67e4e2 / 6 10%–70%0.896497k$0.21 list0
8anthropic/claude-opus-5-5codex · config 3437a9e62 / 6 10%–70%0.926135k$0.60 list0
9fireworks/kimi-k3codex · config 3437a9e62 / 6 10%–70%0.916129k$0.19 list0
10fireworks/qwen3.8-maxopencode · config 357da8912 / 6 10%–70%0.936257k$0.21 list0
11google/gemini-3.8-flashopencode · config 357da8912 / 6 10%–70%0.916572k$0.39 list0
12anthropic/claude-opus-5-5pi · config da41c66e2 / 6 10%–70%0.89657k$0.29 list0
13anthropic/claude-sonnet-5-5pi · config da41c66e2 / 6 10%–70%0.89660k$0.15 list0
14anthropic/claude-opus-5-5claude-code · config c32a950b3 / 13 8%–50%0.886not measured (0/13)unknown16
15openai/gpt-6.1-solcodex · config b23964033 / 15 7%–45%0.906not measured (0/15)unknown12
16google/gemini-3.1-pro @highantigravity · config 034ed3a13 / 18 6%–39%0.866not measured (0/18)unknown0
17zai/glm-5.2claude-code · config ecb67e4e2 / 12 5%–45%0.866not measured (0/12)unknown0
18together/deepseek-v4.1-flashclaude-code · config ecb67e4e1 / 6 3%–56%0.8761.06M$0.33 list0
19anthropic/claude-sonnet-5-5codex · config 3437a9e61 / 6 3%–56%0.836216k$0.47 list0
20moonshot/kimi-k3codex · config 3437a9e61 / 6 3%–56%0.896332k$0.35 list0
21zai/glm-5.2codex · config 3437a9e61 / 6 3%–56%0.856430k$0.33 list0
22anthropic/claude-sonnet-5-5opencode · config 357da8911 / 6 3%–56%0.836380k$0.34 list0
23baseten/glm-5.3-flashopencode · config 357da8911 / 6 3%–56%0.896361k$0.04 list0
24moonshot/kimi-k3opencode · config 357da8911 / 6 3%–56%0.896423k$0.51 list0
25zai/glm-5.2opencode · config 357da8911 / 6 3%–56%0.876479k$0.27 list0
26fireworks/kimi-k3pi · config da41c66e1 / 6 3%–56%0.896217k$0.47 list0
27moonshot/kimi-k3pi · config da41c66e1 / 6 3%–56%0.896174k$0.38 list0
28together/deepseek-v4.1-flashopencode · config 357da8911 / 12 1%–35%0.836607k$0.19 list0
29anthropic/claude-sonnet-5claude-code · config c32a950b0 / 12 0%–24%0.846no passesunknown6
30anthropic/claude-haiku-4-5claude-code · config c32a950b0 / 6 0%–39%0.786no passesunknown6
31baseten/glm-5.3-flashclaude-code · config ecb67e4e0 / 6 0%–39%0.876no passesunknown0
32baseten/glm-5.3-flashcodex · config 3437a9e60 / 6 0%–39%0.826no passesunknown0
33fireworks/qwen3.8-maxcodex · config 3437a9e60 / 6 0%–39%0.806no passesunknown6
34fireworks/kimi-k3opencode · config 357da8910 / 6 0%–39%0.836no passesunknown0
35baseten/glm-5.3-flashpi · config da41c66e0 / 6 0%–39%0.876no passesunknown0
36fireworks/qwen3.8-maxpi · config da41c66e0 / 6 0%–39%0.846no passesunknown0
37moonshot/kimi-k3claude-code · config ecb67e4e0 / 2 0%–66%0.832no passesunknown0
list = published-rate · invoice = provider-invoice-reconciled