| # | Model | Arena score | Samples |
|---|---|---|---|
| 41 | 1,510±6 | 14,946 | |
| 42 |  kimi-k2.6Moonshot | 1,509±7 | 9,585 |
| 43 |  glm-5.1Z.ai | 1,508±7 | 10,220 |
| 44 |  gemini-3.5-flash-highGoogle | 1,500±7 | 8,931 |
| 45 |  claude-opus-4-5-20251101-high-32kAnthropic | 1,495±8 | 10,592 |
| 46 | gemini-3.5-flash-mediumGoogle | 1,492±7 | 11,412 |
| 47 |
| 1,487±6 |
| 17,020 |
| 48 |  minimax-m3MiniMax | 1,487±6 | 12,686 |
|---|
| 49 | qwen3.6-max-previewAlibaba | 1,479±13 | 2,601 |
|---|
| 50 | mimo-v2.5-proXiaomi | 1,475±6 | 15,843 |
|---|
| 51 | kimi-k2.7-codeMoonshot | 1,473±10 | 4,683 |
|---|
| 52 | claude-opus-4-5-20251101Anthropic | 1,468±7 | 12,899 |
|---|
| 53 |  deepseek-v4-pro-high-previewDeepSeek | 1,464±6 | 12,405 |
|---|
| 54 | 1,463±19 | 1,277 |
| 55 | qwen3.6-plusAlibaba | 1,461±6 | 17,562 |
|---|
| 56 |  gpt-5.5 (codex-harness)OpenAI | 1,458±6 | 15,012 |
|---|
| 57 | gemini-3.1-pro-previewGoogle | 1,447±5 | 22,891 |
|---|
| 58 | gemini-3.5-flash-liteGoogle | 1,447±42 | 228 |
|---|
| 59 | deepseek-v4-proDeepSeek | 1,446±6 | 13,183 |
|---|
| 60 | 1,443±19 | 1,316 |
Dots: score · lines: published uncertainty