Mathematics
LiveBench · 28 models
| Rank | Model | Score |
|---|---|---|
| 1 | gpt-5.5-xhigh | 95.8573 |
| 2 | claude-fable-5-xhigh-effort | 95.693 |
| 3 | claude-opus-4-8-xhigh-effort | 95.3165 |
| 4 | gpt-5.5-high | 95.19 |
| 5 | gpt-5.4-xhigh | 94.148 |
| 6 | gpt-5.2-2025-12-11-high | 93.166 |
| 7 | claude-sonnet-5-xhigh-effort | 92.9423 |
| 8 | claude-opus-4-7-xhigh-effort | 92.8537 |
| 9 | gemini-3.1-pro-preview-high | 91.045 |
| 10 | gpt-5.4-nano-xhigh | 90.977 |
| 11 | deepseek-v4-pro | 90.6755 |
| 12 | claude-opus-4-5-20251101-thinking-64k-high-effort | 90.389 |
| 13 | glm-5.2 | 89.7813 |
| 14 | claude-opus-4-6-thinking-auto-high-effort | 89.317 |
| 15 | gpt-5.2-codex | 88.7738 |
| 16 | gemini-3.5-flash-high | 88.2438 |
| 17 | claude-sonnet-4-6-thinking-auto-medium-effort | 86.994 |
| 18 | qwen3.7-max | 85.2483 |
| 19 | grok-4.3 | 84.339 |
| 20 | kimi-k2.6-thinking | 84.2763 |
| 21 | qwen3.6-plus | 83.7248 |
| 22 | qwen3.6-27b | 79.8685 |
| 23 | deepseek-v4-flash | 79.6475 |
| 24 | kimi-k2.7-code | 79.6043 |
| 25 | gpt-5.4-mini-xhigh | 78.462 |
| 26 | grok-build-0.1 | 78.4288 |
| 27 | minimax-m3 | 76.9475 |
| 28 | gpt-5.5 | 69.7695 |