simpleQA
9 models
Top 5 Models Performance
| DeepSeek: deepseek-v3.2-exp | ######################################## | 97.1 |
| SpaceXAI: Grok 4 Fast | ####################################### | 95 |
| DeepSeek: deepseek-v4-pro-0813 | ######################## | 57.9 |
| Tencent: hy3-preview-base | ########### | 26.47 |
| Mistral AI: mistral-small-3.2-24b-instruct-2506 | ##### | 12.1 |
| Rank | Model | Score | |
|---|---|---|---|
| 🇨🇳 | 1 | DeepSeek: deepseek-v3.2-exp深度求索 | 97.1 |
| 🇺🇸 | 2 | SpaceXAI: Grok 4 Fast | 95 |
| 🇨🇳 | 3 | DeepSeek: deepseek-v4-pro-0813深度求索 | 57.9 |
| 🇨🇳 | 4 | Tencent: hy3-preview-base腾讯控股有限公司 | 26.47 |
| 🇫🇷 | 5 | Mistral AI: mistral-small-3.2-24b-instruct-2506 | 12.1 |
| 🇺🇸 | 6 | Google: Gemma 3 27B Instruct | 10 |
| 🇺🇸 | 7 | Google: Gemma 3 12B Instruct | 6.3 |
| 🇺🇸 | 8 | Google: Gemma 3 4B Instruct | 4 |
| 🇺🇸 | 9 | Google: Gemma 3 1B Instruct | 2.2 |