Board · task version 3

review

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.

3 tasks
  • review-corrupt-refresh-checkpoint
  • review-rescue-local-tip
  • review-stale-lock-recovery
review · task version 3
#Liverymodel · harness · configpassed / n95% Wilsonmeantaskstokens / passcost / passexcluded
1openai/gpt-6.1-sol @mediumcodex · config b239640323 / 23 86%–100%1.003107k$0.05 list0
2openai/gpt-6.1-sol @highcodex · config b239640321 / 21 85%–100%1.003108k$0.05 list1
3anthropic/claude-opus-5-5claude-code · config c32a950b12 / 12 76%–100%1.003789k$0.32 list0
4openai/gpt-6.1-solcodex · config b239640312 / 12 76%–100%1.00396k$0.05 list0
5anthropic/claude-opus-5-5 @xhighclaude-code · config c32a950b10 / 10 72%–100%0.9821.26M$0.75 list0
6openai/gpt-6.1-sol @xhighcodex · config b239640310 / 10 72%–100%1.002114k$0.06 list0
7google/gemini-3.8-flashpi · config da41c66e10 / 10 72%–100%1.003397k$0.23 list1
8google/gemini-3.8-flash @highantigravity · config 034ed3a19 / 9 70%–100%1.003827k$0.40 list0
9openai/gpt-6.1-solopencode · config 357da8917 / 7 65%–100%1.003129k$0.09 list0
10google/gemini-3.1-pro @highantigravity · config 034ed3a16 / 6 61%–100%1.003288kunknown0
11anthropic/claude-sonnet-5-5claude-code · config c32a950b6 / 6 61%–100%1.003767k$0.15 list0
12together/deepseek-v4.1-flashopencode · config 357da8915 / 5 57%–100%1.00382k$0.03 list1
13google/gemini-4-argon @highantigravity · config 034ed3a13 / 3 44%–100%1.003249k$0.53 list0
14google/gemini-3.8-flashclaude-code · config ecb67e4e3 / 3 44%–100%1.003802k$0.32 list0
15google/gemini-3.8-flash @mediumclaude-code · config ecb67e4e3 / 3 44%–100%1.003481k$0.32 list0
16openai/gpt-6.1-solclaude-code · config ecb67e4e3 / 3 44%–100%1.003194k$0.09 list0
17google/gemini-3.8-flashcodex · config 3437a9e63 / 3 44%–100%1.003501k$0.27 list0
18google/gemini-3.8-flash @lowcodex · config 3437a9e63 / 3 44%–100%0.933244k$0.15 list0
19anthropic/claude-sonnet-5-5opencode · config 357da8913 / 3 44%–100%0.93377k$0.07 list0
20google/gemini-3.8-flashopencode · config 357da8913 / 3 44%–100%1.003450k$0.22 list0
21google/gemini-3.8-flash @highopencode · config 357da8913 / 3 44%–100%1.003565k$0.32 list0
22anthropic/claude-sonnet-5-5pi · config da41c66e3 / 3 44%–100%1.00326k$0.07 list0
23fireworks/kimi-k3pi · config da41c66e3 / 3 44%–100%1.00367k$0.21 list0
24google/gemini-3.8-flash @highpi · config da41c66e3 / 3 44%–100%1.003693k$0.37 list0
25moonshot/kimi-k3pi · config da41c66e3 / 3 44%–100%1.00359k$0.18 list0
26openai/gpt-6.1-solpi · config da41c66e3 / 3 44%–100%1.00339k$0.04 list0
27together/deepseek-v4.1-flashpi · config da41c66e11 / 12 65%–99%0.973129k$0.05 list1
28together/deepseek-v4.1-flashclaude-code · config ecb67e4e7 / 8 53%–98%0.953351k$0.13 list0
29fireworks/kimi-k3codex · config 3437a9e65 / 7 36%–92%0.833109k$0.21 list0
30fireworks/qwen3.8-maxpi · config da41c66e5 / 7 36%–92%0.863382k$0.31 list0
31baseten/glm-5.3-flashclaude-code · config ecb67e4e2 / 3 21%–94%0.873737k$0.08 list0
32fireworks/kimi-k3claude-code · config ecb67e4e2 / 3 21%–94%0.873481k$0.34 list0
33google/gemini-3.8-flash @highclaude-code · config ecb67e4e2 / 3 21%–94%1.003779k$0.35 list0
34moonshot/kimi-k3claude-code · config ecb67e4e2 / 3 21%–94%0.873385k$0.42 list0
35anthropic/claude-opus-5-5codex · config 3437a9e62 / 3 21%–94%0.87386k$0.41 list0
36baseten/glm-5.3-flashcodex · config 3437a9e62 / 3 21%–94%0.80371k$0.01 list0
37moonshot/kimi-k3codex · config 3437a9e62 / 3 21%–94%0.87392k$0.24 list0
38anthropic/claude-opus-5-5opencode · config 357da8912 / 3 21%–94%0.873100k$0.21 list0
39fireworks/kimi-k3opencode · config 357da8912 / 3 21%–94%0.873156k$0.44 list1
40fireworks/qwen3.8-maxopencode · config 357da8912 / 3 21%–94%0.873192k$0.22 list0
41google/gemini-3.8-flash @lowopencode · config 357da8912 / 3 21%–94%0.873144k$0.10 list0
42moonshot/kimi-k3opencode · config 357da8912 / 3 21%–94%0.873150k$0.32 list0
43anthropic/claude-opus-5-5pi · config da41c66e2 / 3 21%–94%0.87338k$0.21 list0
44fireworks/qwen3.8-maxclaude-code · config ecb67e4e6 / 10 31%–83%0.783467k$0.49 list3
45baseten/glm-5.3-flashopencode · config 357da8914 / 7 25%–84%0.803212k$0.03 list1
46zai/glm-5.2codex · config 3437a9e64 / 8 22%–78%0.803772k$1.08 list0
47fireworks/qwen3.8-27bclaude-code · config ecb67e4e1 / 2 9%–91%0.702323kunknown1
48baseten/glm-5.3-flashpi · config da41c66e3 / 7 16%–75%0.803103k$0.02 list0
49zai/glm-5.2claude-code · config ecb67e4e2 / 6 10%–70%0.7331.52M$1.03 list1
50together/deepseek-v4.1-flashcodex · config 3437a9e62 / 6 10%–70%0.773283k$0.14 list0
51zai/glm-5.2pi · config da41c66e2 / 6 10%–70%0.733636k$0.76 list0
52google/gemini-3.8-flash @lowclaude-code · config ecb67e4e1 / 3 6%–79%0.733358k$0.28 list0
53anthropic/claude-sonnet-5-5codex · config 3437a9e61 / 3 6%–79%0.873149k$0.33 list0
54fireworks/qwen3.8-maxcodex · config 3437a9e61 / 3 6%–79%0.873510k$0.70 list0
55zai/glm-5.2opencode · config 357da8911 / 6 3%–56%0.5731.09M$1.08 list0
56google/gemini-3.8-flash @lowpi · config da41c66e— 0%–0%0not measured (0/0)unknown3
list = published-rate · invoice = provider-invoice-reconciled