评测来源/LiveBench 语言与指令
观察中知识与语言
LiveBench 语言与指令
LiveBench · LiveBench 语言与指令 的公开模型评测结果。
本来源的模型成绩
| Gemini 3.8 Flash | 84.146857 | 84.1 | 2026-06-25 |
| Claude Fable 5 | 82.162143 | 82.2 | 2026-06-25 |
| Gemini 3.1 Pro Preview | 81.789714 | 81.8 | 2026-06-25 |
| GPT-6 Astra | 81.515429 | 81.5 | 2026-06-25 |
| GPT-6.1 Sol | 80.999286 | 81.0 | 2026-06-25 |
| Claude Fable 5.1 | 80.064857 | 80.1 | 2026-06-25 |
| Muse Spark 1.3 | 80.056857 | 80.1 | 2026-06-25 |
| Gemini 3.6 Flash | 79.022857 | 79.0 | 2026-06-25 |
| GPT-5.6 Sol | 78.633571 | 78.6 | 2026-06-25 |
| GPT-5.5 | 77.858 | 77.9 | 2026-06-25 |
| Kimi K3 | 77.433571 | 77.4 | 2026-06-25 |
| Grok 4.7 | 77.359571 | 77.4 | 2026-06-25 |
| Muse Wonder | 77.273429 | 77.3 | 2026-06-25 |
| Grok 4.6 | 76.936714 | 76.9 | 2026-06-25 |
| Qwen3.8 Max | 76.485143 | 76.5 | 2026-06-25 |
| Qwen3.7 Max | 76.483286 | 76.5 | 2026-06-25 |
| Grok 4.5 | 76.357571 | 76.4 | 2026-06-25 |
| Muse Spark 1.2 | 76.146286 | 76.1 | 2026-06-25 |
| Claude Sonnet 5.5 | 76.050714 | 76.1 | 2026-06-25 |
| GPT-6 Sol | 75.742143 | 75.7 | 2026-06-25 |
| GPT-5.4 | 75.538286 | 75.5 | 2026-06-25 |
| Claude Opus 4.8 | 75.301714 | 75.3 | 2026-06-25 |
| Claude Opus 5.5 | 74.956429 | 75.0 | 2026-06-25 |
| DeepSeek V4.1 Flash | 74.814571 | 74.8 | 2026-06-25 |
| Claude Opus 5 | 74.447143 | 74.4 | 2026-06-25 |
| DeepSeek V4 Pro 0813 | 73.860286 | 73.9 | 2026-06-25 |
| GLM 5.3 | 73.826429 | 73.8 | 2026-06-25 |
| GPT-5.6 Terra | 72.449143 | 72.4 | 2026-06-25 |
| Nemotron 3 Ultra 550B A55B | 72.316571 | 72.3 | 2026-06-25 |
| Claude Opus 4.6 | 71.865571 | 71.9 | 2026-06-25 |
| Muse Spark 1.1 | 71.653714 | 71.7 | 2026-06-25 |
| Inkling | 71.537 | 71.5 | 2026-06-25 |
| Claude Opus 4.7 | 71.527429 | 71.5 | 2026-06-25 |
| DeepSeek V4 Flash 0731 | 71.370286 | 71.4 | 2026-06-25 |
| Claude Opus 4.5 | 70.566714 | 70.6 | 2026-06-25 |
| GPT-5.2 | 69.501429 | 69.5 | 2026-06-25 |
| Gemini 3.5 Flash Lite | 69.201714 | 69.2 | 2026-06-25 |
| DeepSeek V4 Pro | 69.113 | 69.1 | 2026-06-25 |
| Kimi K2.6 | 68.979571 | 69.0 | 2026-06-25 |
| Claude Sonnet 4.6 | 68.741571 | 68.7 | 2026-06-25 |
| Claude Sonnet 5 | 68.620714 | 68.6 | 2026-06-25 |
| Grok Build 0.1 | 68.323286 | 68.3 | 2026-06-25 |
| GLM-5.2 | 68.270286 | 68.3 | 2026-06-25 |
| Grok 4.3 | 67.391429 | 67.4 | 2026-06-25 |
| DeepSeek V4 Flash | 66.131714 | 66.1 | 2026-06-25 |
| MiniMax-M3 | 65.791429 | 65.8 | 2026-06-25 |
| Kimi K2.7 Code | 65.554857 | 65.6 | 2026-06-25 |
| Qwen3.6 Plus | 65.476571 | 65.5 | 2026-06-25 |
| GPT-5.6 Luna | 65.455 | 65.5 | 2026-06-25 |
| Claude Haiku 5.5 | 65.202 | 65.2 | 2026-06-25 |
| GPT-5.4 mini | 64.581571 | 64.6 | 2026-06-25 |
| GPT-6 Luna | 63.602286 | 63.6 | 2026-06-25 |
| GLM-5.3 Flash | 63.315 | 63.3 | 2026-06-25 |
| Mistral Large 4 | 58.315714 | 58.3 | 2026-06-25 |
| Qwen3.6 27B | 57.547 | 57.5 | 2026-06-25 |