跳过主要内容
评测来源/LiveBench 数据分析
观察中专业工作

LiveBench 数据分析

LiveBench · LiveBench 数据分析 的公开模型评测结果。

官方评测
来源说明

该来源测量模型在「专业工作」领域的具体交付表现,原始成绩会和统一能力读数分开展示。

限制: 不同版本、推理设置和测试集可能导致分数变化,请结合数据时间阅读。

使用方式

参与排名的来源会影响能力分;参考、交叉参考和观察来源仅用于对比与核验。

本来源的模型成绩

GPT-6 Astra82.97333383.02026-06-25
GPT-6.1 Sol82.65933382.72026-06-25
Muse Wonder81.61733381.62026-06-25
GPT-5.581.57566781.62026-06-25
GPT-6 Sol81.18933381.22026-06-25
Claude Fable 580.53766780.52026-06-25
Claude Opus 5.580.30780.32026-06-25
Claude Fable 5.180.27566780.32026-06-25
GPT-5.6 Sol79.84066779.82026-06-25
Muse Spark 1.379.56533379.62026-06-25
DeepSeek V4 Flash 073179.32733379.32026-06-25
GPT-5.479.31333379.32026-06-25
GPT-5.6 Terra79.30666779.32026-06-25
DeepSeek V4.1 Flash79.25333379.32026-06-25
DeepSeek V4 Pro 081379.24333379.22026-06-25
Kimi K378.73366778.72026-06-25
Claude Sonnet 5.578.6278.62026-06-25
Gemini 3.1 Pro Preview78.54133378.52026-06-25
Qwen3.8 Max78.41378.42026-06-25
Claude Opus 4.778.26366778.32026-06-25
GPT-5.278.16333378.22026-06-25
GPT-5.6 Luna78.03333378.02026-06-25
Claude Sonnet 4.677.94677.92026-06-25
Grok 4.776.88333376.92026-06-25
Mistral Large 476.54166776.52026-06-25
Muse Spark 1.276.45833376.52026-06-25
GLM-5.3 Flash76.40176.42026-06-25
MiniMax-M376.16633376.22026-06-25
Claude Opus 574.55033374.62026-06-25
DeepSeek V4 Pro74.53766774.52026-06-25
Claude Opus 4.574.44166774.42026-06-25
Grok 4.673.85773.92026-06-25
GLM-5.273.73966773.72026-06-25
GPT-6 Luna73.36966773.42026-06-25
Grok 4.573.03766773.02026-06-25
Inkling72.77872.82026-06-25
Claude Haiku 5.572.75372.82026-06-25
Muse Spark 1.172.54872.52026-06-25
Qwen3.7 Max71.78671.82026-06-25
Claude Sonnet 571.74066771.72026-06-25
Grok Build 0.170.79470.82026-06-25
GPT-5.4 mini70.78566770.82026-06-25
Qwen3.6 27B70.42666770.42026-06-25
GLM 5.370.24433370.22026-06-25
Qwen3.6 Plus69.91166769.92026-06-25
Claude Opus 4.669.89369.92026-06-25
DeepSeek V4 Flash68.02368.02026-06-25
Claude Opus 4.866.03333366.02026-06-25
Kimi K2.665.13433365.12026-06-25
Gemini 3.6 Flash62.99933363.02026-06-25
Kimi K2.7 Code62.65733362.72026-06-25
Grok 4.355.77266755.82026-06-25
Nemotron 3 Ultra 550B A55B54.45666754.52026-06-25
Gemini 3.8 Flash54.00566754.02026-06-25
Gemini 3.5 Flash Lite53.24966753.22026-06-25

常见问题