| # | Model | Arena score | Samples |
|---|---|---|---|
| 101 |  gpt-5.2-codexOpenAI | 1,339±9 | 6,402 |
| 102 | gpt-5.1-codexOpenAI | 1,336±12 | 4,861 |
| 103 | 1,331±10 | 5,401 | |
| 104 |  claude-haiku-4-5-20251001Anthropic | 1,329±5 | 27,785 |
| 105 |  deepseek-v3.2DeepSeek | 1,325±8 | 8,966 |
| 106 |  kimi-k2-thinking-turboMoonshot | 1,323±8 | 12,631 |
| 107 |
| 1,302±11 |
| 3,934 |
| 108 |  minimax-m2MiniMax | 1,298±11 | 6,539 |
|---|
| 109 | mimo-v2-flash (thinking)Xiaomi | 1,293±17 | 1,658 |
|---|
| 110 | 1,274±8 | 12,265 |
| 111 | deepseek-v3.2-expDeepSeek | 1,272±14 | 3,865 |
|---|
| 112 | mistral-medium-3.5Mistral | 1,265±15 | 2,154 |
|---|
| 113 | KAT-Coder-Pro-V1Not listed | 1,255±20 | 1,458 |
|---|
| 114 | 1,254±7 | 13,634 |
| 115 | qwen3.5-35b-a3bAlibaba | 1,250±18 | 1,535 |
|---|
| 116 | gpt-5.1-codex-miniOpenAI | 1,244±22 | 1,133 |
|---|
| 117 | grok-4-1-fast-reasoningSpaceXAI | 1,240±11 | 5,314 |
|---|
| 118 | qwen3.5-flashAlibaba | 1,238±20 | 1,270 |
|---|
| 119 | trinity-large-thinkingNot listed | 1,237±21 | 1,303 |
|---|
| 120 | mistral-large-3Mistral | 1,230±26 | 812 |
|---|
Dots: score · lines: published uncertainty
