Benchmarks / Knowledge

SimpleQA Verified (Epoch AI run)

Google DeepMind's cleaned-up SimpleQA short-answer factuality set, run by Epoch AI. Share answered correctly. Data: Epoch AI, CC BY.

80models
–Open Index weight
–last updated
epoch-taskspublisher

SimpleQA Verified (Epoch AI run)

Higher is better · best effort per model

Chart settings
80 of 80 selected

full table below

#ModelEffortScoreCost / taskHarness
1Gemini 4high76.0%–Epoch AI
2GPT-6 Astramax75.6%–Epoch AI
3GPT-6.1 Solmax73.9%–Epoch AI
4Gemini 3.1 Prohigh73.5%–Epoch AI
5Opus 5.5max72.2%–Epoch AI
6Fable 5.1max70.8%–Epoch AI
7Fable 5xhigh70.7%–Epoch AI
8Gemini 3.8 Flashhigh69.7%–Epoch AI
9GPT-5.6 Solmax69.7%–Epoch AI
10Gemini 3.7 Flashhigh69.2%–Epoch AI
11Gemini 3 Flashhigh66.8%–Epoch AI
12Gemini 3.5 Flashhigh66.2%–Epoch AI
13Gemini 3.6 Flashhigh66.2%–Epoch AI
14GPT-5.5xhigh63.0%–Epoch AI
15GPT-6 Solmax60.7%–Epoch AI
16Muse Spark 1.2xhigh60.3%–Epoch AI
17Opus 5max59.9%–Epoch AI
18Muse Spark 1.1max57.8%–Epoch AI
19Grok 4.7xhigh56.0%–Epoch AI
20Qwen 3.7 Maxmax55.8%–Epoch AI
21Opus 4.8max53.0%–Epoch AI
22DeepSeek V4 Pro 0813max52.9%–Epoch AI
23Qwen 3.6 Maxmax52.0%–Epoch AI
24Opus 4.7xhigh51.7%–Epoch AI
25Kimi K3max50.6%–Epoch AI
26GPT-5high50.1%–Epoch AI
27o3high49.4%–Epoch AI
28Grok 4.6high49.3%–Epoch AI
29Qwen 3 Maxmax48.7%–Epoch AI
30Grok 4.5high48.3%–Epoch AI
31GPT-5.1high48.0%–Epoch AI
32Qwen 3.8 Maxxhigh47.3%–Epoch AI
33Opus 4.6max47.0%–Epoch AI
34DeepSeek V4 Promax47.0%–Epoch AI
35Sonnet 5.5max46.5%–Epoch AI
36GPT-5.4 Proxhigh46.3%–Epoch AI
37Opus 4.5high45.7%–Epoch AI
38GPT-5.4xhigh45.1%–Epoch AI
39Qwen 3.6 Plusmax44.1%–Epoch AI
40GPT-5.6 Terramax43.2%–Epoch AI
41DeepSeek V4.1 Flashmax42.9%–Epoch AI
42GPT-6 Lunamax41.4%–Epoch AI
43o1high41.1%–Epoch AI
44GLM 5.3max41.0%–Epoch AI
45GPT-5.6 Lunamax41.0%–Epoch AI
46Qwen 3 235B-A22Bxhigh40.4%–Epoch AI
47Inklingxhigh40.3%–Epoch AI
48GPT-5.2xhigh37.1%–Epoch AI
49Kimi K2.7 Codemax36.5%–Epoch AI
50Sonnet 4.6high35.5%–Epoch AI
51Kimi K2.6max34.9%–Epoch AI
52Kimi K2.5max34.3%–Epoch AI
53GLM 5.2max34.2%–Epoch AI
54GLM 5.1max34.0%–Epoch AI
55Sonnet 5max33.7%–Epoch AI
56DeepSeek V4 Flash 0731max33.6%–Epoch AI
57Grok 4.3high33.2%–Epoch AI
58GLM 4.7none32.2%–Epoch AI
59GPT-4.1high31.1%–Epoch AI
60Sonnet 4.5high30.7%–Epoch AI
61Grok 4.20xhigh30.2%–Epoch AI
62GPT-5.4 Minihigh29.4%–Epoch AI
63GPT-4ohigh26.0%–Epoch AI
64Qwen 3.5 Plusnone25.4%–Epoch AI
65Haiku 5.5max23.8%–Epoch AI
66GPT-5 Minihigh21.6%–Epoch AI
67Qwen 3.5 Flashmax20.3%–Epoch AI
68Mistral Large 4high20.0%–Epoch AI
69o4 minilow19.6%–Epoch AI
70Inkling Smallxhigh19.1%–Epoch AI
71Qwen 3.6 Flashmax15.9%–Epoch AI
72o3 minihigh15.3%–Epoch AI
73GPT-4.1 Minihigh12.7%–Epoch AI
74Opus 3none12.6%–Epoch AI
75Haiku 4.5xhigh12.6%–Epoch AI
76GPT-5.4 Nanohigh11.7%–Epoch AI
77GPT-5 Nanohigh11.7%–Epoch AI
78Gemma 4 31B IThigh10.4%–Epoch AI
79GPT-4o minihigh8.3%–Epoch AI
80GPT-4.1 Nanohigh6.0%–Epoch AI