SimpleQA Verified
Short factual questions; measures recall without hallucination. Run by Epoch AI.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Knowledge
- Index weight
- 1.0
- Models
- 75
- Data as of
- 9 Sept 2026
CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.
- 1GPT-6 Astra75.6%
- 2Gemini 3.1 Pro73.5%
- 3Claude Fable 5.170.8%
- 4Claude Fable 570.7%
- 5GPT-5.6 Sol69.7%
- 6Gemini 3.8 Flash69.7%
- 7Gemini 3.7 Flash69.2%
- 8Gemini 3 Flash66.8%
- 9Gemini 3.5 Flash66.2%
- 10Gemini 3.6 Flash66.2%
- 11GPT-5.563.0%
- 12Muse Spark 1.260.3%
- 13Claude Opus 559.9%
- 14Muse Spark 1.157.8%
- 15Qwen3 755.8%
75 of 75
| # | |||||
|---|---|---|---|---|---|
| 1 |  GPT-6 AstraOpenAI | 75.6% | – | 70.6 | – |
| 2 |  Gemini 3.1 ProGoogle | 73.5% | – | 60.7 | – |
| 3 |  Claude Fable 5.1Anthropic | 70.8% | – | 70.2 | – |
| 4 | Claude Fable 5Anthropic | 70.7% | – | – | – |
| 5 | GPT-5.6 SolOpenAI | 69.7% | – | 65.5 | – |
| 6 | Gemini 3.8 FlashGoogle | 69.7% | – | 60.1 | – |
| 7 | Gemini 3.7 FlashGoogle | 69.2% | – | 60.8 | – |
| 8 | Gemini 3 FlashGoogle | 66.8% | – | 58.5 | – |
| 9 | Gemini 3.5 FlashGoogle | 66.2% | – | 60.4 | – |
| 10 | Gemini 3.6 FlashGoogle | 66.2% | – | 58.6 | – |
| 11 | GPT-5.5OpenAI | 63.0% | – | 63.7 | – |
| 12 |  Muse Spark 1.2Meta | 60.3% | – | 61.3 | – |
| 13 | Claude Opus 5Anthropic | 59.9% | – | 66.8 | – |
| 14 | Muse Spark 1.1Meta | 57.8% | – | 64.9 | – |
| 15 |  Qwen3 7Alibaba | 55.8% | – | 58.9 | – |
| 16 | Claude Opus 4.8Anthropic | 53.0% | – | 59.4 | – |
| 17 |  DeepSeek V4 ProDeepSeekopen | 52.9% | max, 0813 | 56.1 | – |
| 18 | Qwen3 6Alibaba | 52.0% | – | 59.5 | – |
| 19 | Claude Opus 4.7Anthropic | 51.7% | – | 57.2 | – |
| 20 |  Kimi K3Moonshot AIopen | 50.6% | – | 62.7 | – |
| 21 | GPT-5OpenAI | 50.1% | – | 54.5 | – |
| 22 | o3OpenAI | 49.4% | – | 52.9 | – |
| 23 |  Grok 4.6xAI | 49.3% | – | 58.7 | – |
| 24 | Grok 4.6xAI | 48.9% | – | 62.3 | – |
| 25 | Qwen3 MaxAlibaba | 48.8% | – | 52.7 | – |
| 26 | Grok 4.5xAI | 48.3% | – | 54.2 | – |
| 27 | GPT-5.1OpenAI | 48.0% | – | 53.8 | – |
| 28 | Qwen3 8Alibaba | 47.3% | xhigh, 0902 | 58.0 | – |
| 29 | Claude Opus 4.6Anthropic | 47.0% | – | 54.1 | – |
| 30 | GPT-5.4 ProOpenAI | 46.3% | – | 61.1 | – |
| 31 | Claude Opus 4.5Anthropic | 45.7% | – | 58.0 | – |
| 32 | GPT-5.4OpenAI | 45.1% | – | 61.3 | – |
| 33 | Qwen3.6 PlusAlibabaopen | 44.1% | – | 59.0 | – |
| 34 | GPT-5.6 TerraOpenAI | 43.2% | – | 58.0 | – |
| 35 | o1OpenAI | 41.1% | – | 44.3 | – |
| 36 |  GLM-5.3Zhipu AIopen | 41.0% | – | 59.9 | – |
| 37 | GPT-5.6 LunaOpenAI | 41.0% | – | 56.5 | – |
| 38 | Qwen3 235B A22BAlibabaopen | 40.4% | thinking, 2507 | 51.7 | – |
| 39 | GPT-5.2OpenAI | 37.1% | – | 57.3 | – |
| 40 | Kimi K2.7 CodeMoonshot AIopen | 36.5% | – | 56.2 | – |
| 41 | Claude Sonnet 4.6Anthropic | 35.5% | – | 52.3 | – |
| 42 | Kimi K2.6Moonshot AIopen | 34.9% | kimi-k2.6 | 60.6 | – |
| 43 | Kimi K2.5Moonshot AIopen | 34.3% | kimi-k2.5 | 56.1 | – |
| 44 | GPT-5.2OpenAI | 34.3% | – | 55.5 | – |
| 45 | GLM-5.2Zhipu AIopen | 34.2% | – | 57.6 | – |
| 46 | GLM-5.1Zhipu AIopen | 34.0% | glm-5.1 | 60.1 | – |
| 47 | Claude Sonnet 5Anthropic | 33.7% | – | 50.7 | – |
| 48 | DeepSeek V4 FlashDeepSeekopen | 33.6% | max, 0731 | 54.7 | – |
| 49 | Grok 4.3xAI | 33.2% | – | 50.1 | – |
| 50 | Claude Sonnet 5Anthropic | 32.9% | – | 57.3 | – |
| 51 | GPT-5.2OpenAI | 32.8% | – | 50.2 | – |
| 52 | Claude Sonnet 4.6Anthropic | 32.8% | – | 49.7 | – |
| 53 | GPT-5.2OpenAI | 32.7% | – | 58.5 | – |
| 54 | GLM-4.7Zhipu AIopen | 32.2% | glm-4.7 | 54.9 | – |
| 55 | GPT-4.1OpenAI | 31.1% | – | 47.3 | – |
| 56 | Claude Sonnet 4.5Anthropic | 30.7% | – | 52.3 | – |
| 57 | Grok 4.20xAI | 30.2% | – | 54.3 | – |
| 58 | GPT-5.4 miniOpenAI | 29.4% | – | 52.9 | – |
| 59 | GPT-4oOpenAI | 26.0% | – | 42.3 | – |
| 60 | Qwen3.5 PlusAlibaba | 25.4% | – | – | – |