Agentic Coding
LiveBench · 28 models
| Rank | Model | Score |
|---|---|---|
| 1 | claude-opus-4-8-xhigh-effort | 56.0607 |
| 2 | gpt-5.4-xhigh | 53.8383 |
| 3 | gpt-5.5-xhigh | 52.0707 |
| 4 | glm-5.2 | 51.9193 |
| 5 | claude-sonnet-5-xhigh-effort | 51.0607 |
| 6 | claude-fable-5-xhigh-effort | 50.6563 |
| 6 | claude-opus-4-7-xhigh-effort | 50.6563 |
| 8 | gpt-5.2-2025-12-11-high | 50.2527 |
| 9 | gpt-5.2-codex | 49.394 |
| 10 | claude-opus-4-6-thinking-auto-high-effort | 48.9897 |
| 10 | gemini-3.5-flash-high | 48.9897 |
| 12 | kimi-k2.6-thinking | 46.9193 |
| 13 | gpt-5.4-nano-xhigh | 46.7677 |
| 13 | gpt-5.5-high | 46.7677 |
| 15 | grok-build-0.1 | 45.808 |
| 16 | kimi-k2.7-code | 45.6563 |
| 17 | gemini-3.1-pro-preview-high | 45.404 |
| 18 | qwen3.7-max | 43.586 |
| 19 | claude-sonnet-4-6-thinking-auto-medium-effort | 42.626 |
| 19 | deepseek-v4-pro | 42.626 |
| 21 | gpt-5.4-mini-xhigh | 41.6667 |
| 22 | qwen3.6-plus | 41.3637 |
| 23 | gpt-5.5 | 41.111 |
| 24 | minimax-m3 | 40.6563 |
| 25 | claude-opus-4-5-20251101-thinking-64k-high-effort | 39.697 |
| 26 | qwen3.6-27b | 39.2927 |
| 27 | deepseek-v4-flash | 37.626 |
| 28 | grok-4.3 | 18.5353 |