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

Gemini 4 Argon0.7677.62026-10-17
Gemini 4 Argon0.7677.62026-10-18
Gemini 4 Argon0.7677.62026-11-05
GPT-6 Astra0.75677.22026-11-05
GPT-6.1 Sol0.73975.72026-11-05
Gemini 3.1 Pro Preview0.73575.42026-11-05
Claude Opus 5.50.72274.32026-11-05
Claude Fable 5.10.70873.12026-11-05
Claude Fable 50.70773.02026-11-05
GPT-5.6 Sol0.69772.12026-11-05
Gemini 3.8 Flash0.69772.12026-11-05
Gemini 3 Flash Preview0.66869.62026-11-05
Gemini 3.6 Flash0.66269.12026-11-05
GPT-5.50.6366.32026-11-05
GPT-6 Sol0.60764.32026-11-05
Muse Spark 1.20.6030150864.02026-11-05
Claude Opus 50.59963.62026-11-05
Muse Spark 1.10.5778894561.82026-11-05
GPT-6.1 Terra0.57361.42026-11-05
Grok 4.70.5660.32026-11-05
Qwen3.7 Max0.5576730260.12026-11-05
Claude Opus 4.80.5357.72026-11-05
DeepSeek V4 Pro 08130.5290581257.62026-11-05
Qwen3.6 Max Preview0.5200400856.82026-11-05
Claude Opus 4.70.51756.52026-11-05
Kimi K30.50655.62026-11-05
GPT-50.50155.22026-11-05
o30.49454.62026-11-05
Grok 4.60.49354.52026-11-05
Qwen3 Max0.4874874954.02026-11-05
Grok 4.50.48353.62026-11-05
Claude Opus 4.60.4752.52026-11-05
Claude Sonnet 5.50.46552.02026-11-05
GPT-5.4 Pro0.46351.92026-11-05
Qwen3.8 Max0.45851.42026-11-05
Claude Opus 4.50.45751.42026-11-05
GPT-5.40.45150.82026-11-05
Qwen3.6 Plus0.4414414450.02026-11-05
GPT-5.6 Terra0.43249.22026-11-05
DeepSeek V4.1 Flash0.42948.92026-11-05
DeepSeek V4.1 Flash0.42948.92026-10-18
DeepSeek V4.1 Flash0.42948.92026-10-17
DeepSeek V4.1 Flash 10170.42548.62026-11-05
GPT-6 Luna0.41447.62026-11-05
o10.41147.42026-11-03
GPT-5.6 Luna0.4147.32026-11-05
GLM 5.30.4147.32026-11-05
Inkling0.40346.72026-11-05
GPT-5.20.37143.92026-11-05
Kimi K2.7 Code0.36543.42026-11-05
Claude Sonnet 4.60.35542.52026-11-05
Kimi K2.60.34942.02026-11-05
Kimi K2.50.34341.52026-11-05
GLM-5.20.34241.42026-11-05
GLM-5.10.3441.22026-11-05
Claude Sonnet 50.33741.02026-11-05
DeepSeek V4 Flash 07310.3363363440.92026-11-05
GLM-4.70.32239.72026-11-05
Claude Sonnet 4.50.30738.42026-11-05
Grok 4.200.30237.92026-11-05
GPT-5.4 mini0.29437.32026-11-05
Qwen3.5 Plus0.2535070133.72026-11-05
Claude Haiku 5.50.23832.42026-11-05
Mistral Large 40.22431.22026-11-05
GPT-5 mini0.21630.52026-11-05
Qwen3.5-Flash0.203203229.42026-11-05
o4-mini0.19628.82026-11-05
Inkling Small0.19128.32026-11-05
o4 Mini High0.18828.12026-11-05
Qwen3.6 Flash0.1591591625.62026-11-05
o3 Mini High0.15325.12026-11-05
Claude Haiku 4.50.13223.22026-11-05
GPT-5 nano0.11721.92026-11-05
Gemma 4 31B IT0.10420.82026-11-05

常见问题