Table Reformat
LiveBench · 28 models
| Rank | Model | Score |
|---|---|---|
| 1 | deepseek-v4-pro | 100 |
| 1 | gpt-5.2-2025-12-11-high | 100 |
| 1 | gpt-5.2-codex | 100 |
| 1 | gpt-5.4-mini-xhigh | 100 |
| 1 | gpt-5.4-xhigh | 100 |
| 1 | gpt-5.5 | 100 |
| 1 | gpt-5.5-high | 100 |
| 1 | gpt-5.5-xhigh | 100 |
| 1 | grok-build-0.1 | 100 |
| 10 | claude-fable-5-xhigh-effort | 98.039 |
| 10 | claude-opus-4-5-20251101-thinking-64k-high-effort | 98.039 |
| 10 | claude-opus-4-6-thinking-auto-high-effort | 98.039 |
| 10 | claude-opus-4-7-xhigh-effort | 98.039 |
| 10 | claude-opus-4-8-xhigh-effort | 98.039 |
| 10 | claude-sonnet-5-xhigh-effort | 98.039 |
| 10 | deepseek-v4-flash | 98.039 |
| 10 | gemini-3.1-pro-preview-high | 98.039 |
| 10 | gpt-5.4-nano-xhigh | 98.039 |
| 10 | grok-4.3 | 98.039 |
| 10 | kimi-k2.6-thinking | 98.039 |
| 10 | minimax-m3 | 98.039 |
| 10 | qwen3.6-27b | 98.039 |
| 10 | qwen3.6-plus | 98.039 |
| 10 | qwen3.7-max | 98.039 |
| 25 | claude-sonnet-4-6-thinking-auto-medium-effort | 96.078 |
| 25 | gemini-3.5-flash-high | 96.078 |
| 25 | glm-5.2 | 96.078 |
| 28 | kimi-k2.7-code | 92.157 |