跳过主要内容
评测来源/SimpleQA Verified
参与排名知识与语言

SimpleQA Verified

Epoch AI · SimpleQA Verified 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「知识与语言」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

Claude Fable 5.20.9489.62026-11-10
Gemini 4 Carbon0.77776.32026-11-11
Gemini 4 Argon0.7674.92026-10-17
Gemini 4 Argon0.7674.92026-10-18
Gemini 4 Argon0.7674.92026-11-05
GPT-6 Astra0.75674.62026-11-05
GPT-6.1 Sol0.73973.22026-11-05
Gemini 3.1 Pro Preview0.73572.92026-11-05
Claude Opus 5.50.72271.82026-11-05
Claude Fable 5.10.70870.72026-11-05
Claude Fable 50.70770.62026-11-05
GPT-5.6 Sol0.69769.82026-11-05
Gemini 3.8 Flash0.69769.82026-11-05
Gemini 3 Flash Preview0.66867.52026-11-05
Gemini 3.6 Flash0.66267.02026-11-05
GPT-5.50.6364.42026-11-05
GPT-6 Sol0.60762.52026-11-05
Muse Spark 1.20.6030150862.22026-11-05
Claude Opus 50.59961.92026-11-05
Muse Spark 1.10.5778894560.12026-11-05
GPT-6.1 Terra0.57359.72026-11-05
Grok 4.70.5658.72026-11-05
Qwen3.7 Max0.5576730258.52026-11-05
Claude Opus 4.80.5356.32026-11-05
DeepSeek V4 Pro 08130.5290581256.22026-11-05
Qwen3.6 Max Preview0.5200400855.42026-11-05
Claude Opus 4.70.51755.22026-11-05
Kimi K30.50654.32026-11-05
GPT-50.50153.92026-11-05
o30.49453.32026-11-05
Grok 4.60.49353.32026-11-05
Qwen3 Max0.4874874952.82026-11-05
Grok 4.50.48352.42026-11-05
Claude Opus 4.60.4751.42026-11-05
Claude Sonnet 5.50.46551.02026-11-05
GPT-5.4 Pro0.46350.82026-11-05
Qwen3.8 Max0.45850.42026-11-05
Claude Opus 4.50.45750.32026-11-05
GPT-5.40.45149.82026-11-05
Qwen3.6 Plus0.4414414449.12026-11-05
GPT-5.6 Terra0.43248.32026-11-05
DeepSeek V4.1 Flash0.42948.12026-11-05
DeepSeek V4.1 Flash0.42948.12026-10-18
DeepSeek V4.1 Flash0.42948.12026-10-17
DeepSeek V4.1 Flash 10170.42547.72026-11-05
GPT-6 Luna0.41446.82026-11-05
o10.41146.62026-11-03
GPT-5.6 Luna0.4146.52026-11-05
GLM 5.30.4146.52026-11-05
Inkling0.40345.92026-11-05
GPT-5.20.37143.32026-11-05
Kimi K2.7 Code0.36542.92026-11-05
Claude Sonnet 4.60.35542.02026-11-05
Kimi K2.60.34941.62026-11-05
Kimi K2.50.34341.12026-11-05
GLM-5.20.34241.02026-11-05
GLM-5.10.3440.82026-11-05
Claude Sonnet 50.33740.62026-11-05
DeepSeek V4 Flash 07310.3363363440.52026-11-05
GLM-4.70.32239.42026-11-05
Claude Sonnet 4.50.30738.12026-11-05
Grok 4.200.30237.72026-11-05
GPT-5.4 mini0.29437.12026-11-05
Qwen3.5 Plus0.2535070133.82026-11-05
Claude Haiku 5.50.23832.52026-11-05
Mistral Large 40.22431.42026-11-05
GPT-5 mini0.21630.82026-11-05
Qwen3.5-Flash0.203203229.72026-11-05
o4-mini0.19629.12026-11-05
Inkling Small0.19128.72026-11-05
o4 Mini High0.18828.52026-11-05
Qwen3.6 Flash0.1591591626.12026-11-05
o3 Mini High0.15325.62026-11-05
Claude Haiku 4.50.13223.92026-11-05
GPT-5 nano0.11722.72026-11-05
Gemma 4 31B IT0.10421.72026-11-05

常见问题