Paraphrase
LiveBench · 28 models
| Rank | Model | Score |
|---|---|---|
| 1 | gemini-3.1-pro-preview-high | 79.867 |
| 2 | gemini-3.5-flash-high | 75.517 |
| 3 | gpt-5.5-high | 72.117 |
| 4 | qwen3.7-max | 72 |
| 5 | gpt-5.5-xhigh | 71.733 |
| 6 | claude-fable-5-xhigh-effort | 70.183 |
| 7 | claude-opus-4-8-xhigh-effort | 69.917 |
| 8 | gpt-5.2-codex | 67.05 |
| 9 | gpt-5.4-nano-xhigh | 66.649 |
| 10 | claude-opus-4-7-xhigh-effort | 65.767 |
| 11 | claude-opus-4-5-20251101-thinking-64k-high-effort | 65.667 |
| 12 | grok-4.3 | 64.383 |
| 13 | gpt-5.2-2025-12-11-high | 64.1 |
| 14 | gpt-5.5 | 63.6 |
| 15 | gpt-5.4-xhigh | 63.433 |
| 16 | claude-opus-4-6-thinking-auto-high-effort | 62.817 |
| 17 | deepseek-v4-flash | 62.383 |
| 18 | glm-5.2 | 62.25 |
| 19 | kimi-k2.6-thinking | 61.633 |
| 20 | claude-sonnet-5-xhigh-effort | 61.25 |
| 21 | qwen3.6-plus | 60.617 |
| 22 | claude-sonnet-4-6-thinking-auto-medium-effort | 59.4 |
| 23 | gpt-5.4-mini-xhigh | 58.883 |
| 24 | grok-build-0.1 | 58.383 |
| 25 | deepseek-v4-pro | 56.217 |
| 26 | minimax-m3 | 55.933 |
| 27 | qwen3.6-27b | 49.433 |
| 28 | kimi-k2.7-code | 48.533 |