Benchmark
SimpleQA
general
reasoning
text
SimpleQA is a factuality benchmark developed by OpenAI that measures the short-form factual accuracy of large language models. The benchmark contains 4,326 short, fact-seeking questions that are adversarially collected and designed to have single, indisputable answers. Questions cover diverse topics from science and technology to entertainment, and the benchmark also measures model calibration by evaluating whether models know what they know.
语言EN
满分1
参评模型26
模型排名
| 名次 | 模型 | 机构 | 分数 | 来源 |
|---|---|---|---|---|
| 1 | DeepSeek-V3.2-Exp | DeepSeek | 97.1 | 来源 ↗ |
| 2 | Grok 4 Fast | xAI | 95.0 | 来源 ↗ |
| 3 | DeepSeek-V3.1 | DeepSeek | 93.4 | 来源 ↗ |
| 4 | DeepSeek-R1-0528 | DeepSeek | 92.3 | 来源 ↗ |
| 5 | GPT-4.5 | OpenAI | 62.5 | 来源 ↗ |
| 6 | Qwen3-235B-A22B-Instruct-2507 | Alibaba Cloud / Qwen Team | 54.3 | 来源 ↗ |
| 7 | Gemini 2.5 Pro Preview 06-05 | 54.0 | 来源 ↗ | |
| 8 | Gemini 2.5 Pro | 50.8 | 来源 ↗ | |
| 9 | o1 | OpenAI | 47.0 | 来源 ↗ |
| 10 | o1-preview | OpenAI | 42.4 | 来源 ↗ |
| 11 | GPT-4o | OpenAI | 38.2 | 来源 ↗ |
| 12 | Kimi K2 Base | Moonshot AI | 35.3 | 来源 ↗ |
| 13 | Kimi K2-Instruct-0905 | Moonshot AI | 31.0 | 来源 ↗ |
| 14 | Kimi K2 Instruct | Moonshot AI | 31.0 | 来源 ↗ |
| 15 | Gemini 2.5 Flash | 26.9 | 来源 ↗ | |
| 16 | DeepSeek-V3 | DeepSeek | 24.9 | 来源 ↗ |
| 17 | Gemini 2.0 Flash-Lite | 21.7 | 来源 ↗ | |
| 18 | o3-mini | OpenAI | 15.0 | 来源 ↗ |
| 19 | Mistral Small 3.2 24B Instruct | Mistral AI | 12.1 | 来源 ↗ |
| 20 | Gemini 2.5 Flash-Lite | 10.7 | 来源 ↗ | |
| 21 | Mistral Small 3.1 24B Instruct | Mistral AI | 10.4 | 来源 ↗ |
| 22 | Gemma 3 27B | 10.0 | 来源 ↗ | |
| 23 | Gemma 3 12B | 6.3 | 来源 ↗ | |
| 24 | Gemma 3 4B | 4.0 | 来源 ↗ | |
| 25 | Phi 4 | Microsoft | 3.0 | 来源 ↗ |
| 26 | Gemma 3 1B | 2.2 | 来源 ↗ |