Board · task version 1

injection

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.

2 tasks
  • injection-exfil-via-telegram
  • injection-memory-write-injection
injection
#Liverymodel · harness · configpassed / n95% Wilsonmeantaskstokens / passcost / passexcluded
1google/gemini-3.1-pro @highantigravity · config 034ed3a1— 0%–0%0not measured (0/0)unknown2
2google/gemini-4-argon @highantigravity · config 034ed3a1— 0%–0%0not measured (0/0)unknown2
3anthropic/claude-haiku-4-5claude-code · config c32a950b— 0%–0%0not measured (0/0)unknown2
4anthropic/claude-opus-5-5claude-code · config c32a950b— 0%–0%0not measured (0/0)unknown2
5anthropic/claude-sonnet-5claude-code · config c32a950b— 0%–0%0not measured (0/0)unknown2
6anthropic/claude-sonnet-5-5claude-code · config c32a950b— 0%–0%0not measured (0/0)unknown2
7baseten/glm-5.3-flashclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
8fireworks/kimi-k3claude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
9fireworks/qwen3.8-maxclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
10google/gemini-3.8-flashclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
11google/gemini-3.8-flash @highclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
12google/gemini-3.8-flash @lowclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
13google/gemini-3.8-flash @mediumclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
14openai/gpt-6.1-solclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
15together/deepseek-v4.1-flashclaude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
16zai/glm-5.2claude-code · config ecb67e4e— 0%–0%0not measured (0/0)unknown2
17anthropic/claude-opus-5-5codex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
18anthropic/claude-sonnet-5-5codex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
19baseten/glm-5.3-flashcodex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
20fireworks/kimi-k3codex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
21fireworks/qwen3.8-maxcodex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
22google/gemini-3.8-flashcodex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
23google/gemini-3.8-flash @lowcodex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
24moonshot/kimi-k3codex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
25together/deepseek-v4.1-flashcodex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
26zai/glm-5.2codex · config 3437a9e6— 0%–0%0not measured (0/0)unknown2
27openai/gpt-6.1-solcodex · config b2396403— 0%–0%0not measured (0/0)unknown2
28openai/gpt-6.1-sol @highcodex · config b2396403— 0%–0%0not measured (0/0)unknown2
29openai/gpt-6.1-sol @mediumcodex · config b2396403— 0%–0%0not measured (0/0)unknown2
30anthropic/claude-opus-5-5opencode · config 357da891— 0%–0%0not measured (0/0)unknown2
31anthropic/claude-sonnet-5-5opencode · config 357da891— 0%–0%0not measured (0/0)unknown2
32baseten/glm-5.3-flashopencode · config 357da891— 0%–0%0not measured (0/0)unknown2
33fireworks/kimi-k3opencode · config 357da891— 0%–0%0not measured (0/0)unknown2
34fireworks/qwen3.8-maxopencode · config 357da891— 0%–0%0not measured (0/0)unknown2
35google/gemini-3.8-flashopencode · config 357da891— 0%–0%0not measured (0/0)unknown2
36google/gemini-3.8-flash @highopencode · config 357da891— 0%–0%0not measured (0/0)unknown2
37google/gemini-3.8-flash @lowopencode · config 357da891— 0%–0%0not measured (0/0)unknown2
38moonshot/kimi-k3opencode · config 357da891— 0%–0%0not measured (0/0)unknown2
39openai/gpt-6.1-solopencode · config 357da891— 0%–0%0not measured (0/0)unknown2
40together/deepseek-v4.1-flashopencode · config 357da891— 0%–0%0not measured (0/0)unknown4
41zai/glm-5.2opencode · config 357da891— 0%–0%0not measured (0/0)unknown2
42anthropic/claude-opus-5-5pi · config da41c66e— 0%–0%0not measured (0/0)unknown2
43anthropic/claude-sonnet-5-5pi · config da41c66e— 0%–0%0not measured (0/0)unknown2
44baseten/glm-5.3-flashpi · config da41c66e— 0%–0%0not measured (0/0)unknown2
45fireworks/kimi-k3pi · config da41c66e— 0%–0%0not measured (0/0)unknown2
46fireworks/qwen3.8-maxpi · config da41c66e— 0%–0%0not measured (0/0)unknown2
47google/gemini-3.8-flashpi · config da41c66e— 0%–0%0not measured (0/0)unknown2
48google/gemini-3.8-flash @highpi · config da41c66e— 0%–0%0not measured (0/0)unknown2
49google/gemini-3.8-flash @lowpi · config da41c66e— 0%–0%0not measured (0/0)unknown2
50moonshot/kimi-k3pi · config da41c66e— 0%–0%0not measured (0/0)unknown2
51openai/gpt-6.1-solpi · config da41c66e— 0%–0%0not measured (0/0)unknown2
52together/deepseek-v4.1-flashpi · config da41c66e— 0%–0%0not measured (0/0)unknown2
53zai/glm-5.2pi · config da41c66e— 0%–0%0not measured (0/0)unknown2
list = published-rate · invoice = provider-invoice-reconciled