Logic With Navigation
LiveBench · 28 models
| Rank | Model | Score |
|---|---|---|
| 1 | claude-sonnet-5-xhigh-effort | 86 |
| 2 | qwen3.7-max | 84 |
| 3 | claude-opus-4-8-xhigh-effort | 82 |
| 4 | claude-opus-4-6-thinking-auto-high-effort | 80 |
| 4 | glm-5.2 | 80 |
| 6 | gpt-5.5-high | 78 |
| 7 | gpt-5.5 | 76 |
| 7 | gpt-5.5-xhigh | 76 |
| 9 | gemini-3.5-flash-high | 74 |
| 9 | gpt-5.2-2025-12-11-high | 74 |
| 9 | kimi-k2.7-code | 74 |
| 12 | claude-fable-5-xhigh-effort | 72 |
| 12 | gemini-3.1-pro-preview-high | 72 |
| 14 | claude-opus-4-7-xhigh-effort | 70 |
| 14 | claude-sonnet-4-6-thinking-auto-medium-effort | 70 |
| 14 | grok-build-0.1 | 70 |
| 14 | kimi-k2.6-thinking | 70 |
| 14 | qwen3.6-plus | 70 |
| 19 | claude-opus-4-5-20251101-thinking-64k-high-effort | 68 |
| 19 | gpt-5.2-codex | 68 |
| 19 | gpt-5.4-xhigh | 68 |
| 22 | grok-4.3 | 66 |
| 22 | minimax-m3 | 66 |
| 24 | deepseek-v4-flash | 64 |
| 24 | deepseek-v4-pro | 64 |
| 24 | gpt-5.4-nano-xhigh | 64 |
| 27 | qwen3.6-27b | 62 |
| 28 | gpt-5.4-mini-xhigh | 60 |