| # | Model | Arena score | Samples |
|---|---|---|---|
| 81 | qwen3.5-397b-a17bAlibaba | 1,442±3 | 76,643 |
| 82 |  minimax-m3MiniMax | 1,442±4 | 48,130 |
| 83 |  glm-4.7Z.ai | 1,442±6 | 11,892 |
| 84 | inklingThinky | 1,439±5 | 25,509 |
| 85 |  gpt-5.1OpenAI | 1,439±4 | 42,992 |
| 86 |  deepseek-v4-flash-high-previewDeepSeek | 1,438±4 | 48,595 |
| 87 |  |
| 1,438±8 |
| 5,804 |
| 88 | qwen3.8-27bAlibaba | 1,438±7 | 10,256 |
|---|
| 89 | gpt-5.2-highOpenAI | 1,438±4 | 47,539 |
|---|
| 90 | gpt-5.2OpenAI | 1,436±3 | 78,963 |
|---|
| 91 | deepseek-v4-flashDeepSeek | 1,436±4 | 48,890 |
|---|
| 92 | longcat-flash-chat-2602-expMeituan | 1,436±5 | 27,961 |
|---|
| 93 | qwen3-max-previewAlibaba | 1,435±5 | 27,194 |
|---|
| 94 | gpt-5-highOpenAI | 1,435±5 | 31,424 |
|---|
| 95 | mimo-v2.5Xiaomi | 1,433±4 | 44,446 |
|---|
| 96 | glm-5v-turboZ.ai | 1,433±7 | 9,367 |
|---|
| 97 | 1,432±4 | 60,409 |
| 98 | o3-2025-04-16OpenAI | 1,431±4 | 58,569 |
|---|
| 99 |  kimi-k2.5-instantMoonshot | 1,431±7 | 7,990 |
|---|
| 100 | grok-4-1-fast-reasoningSpaceXAI | 1,430±3 | 55,355 |
|---|
Dots: score · lines: published uncertainty