TruthfulQA evaluates tendency to avoid common misconceptions and answer factually when faced with misleading questions.
Models ranked by score on this benchmark. Higher is better.
| 3 |
| OpenAI |
| 72.2 |
| p95 |
| Text Generation, Small, Multimodal, Reasoning, Proprietary |
| 4 | Alibaba | 71.4 | — | Reasoning, Large, Code Assistant, Text Generation, Multimodal, Open Weight |
| 5 | Anthropic | 71.2 | — | Multimodal, Small, Text Generation, Proprietary |
| 6 | Anthropic | 71.0 | p94 | Code Assistant, Small, Text Generation, Multimodal, Reasoning, Proprietary |
| 7 | 69.8 | p93 | Multimodal, Small, Text Generation, Reasoning, Proprietary |
| 8 | Anthropic | 69.0 | p93 | Multimodal, Small, Text Generation, Reasoning, Proprietary, Large |
| 9 | DeepSeek | 68.0 | p92 | Text Generation, Reasoning, Open Weight, Large |
| 10 | Alibaba | 67.0 | p91 | Reasoning, Large, Code Assistant, Text Generation, Multimodal, Open Weight |
| 11 | Meta | 65.2 | p89 | Reasoning, Large, Text Generation, Open Weight |
| 12 | Alibaba | 64.5 | p88 | Text Generation, Large, Multimodal, Reasoning, Open Weight |
| 13 | Mistral AI | 63.8 | p88 | Text Generation, Small, Reasoning, Proprietary |
| 14 | Meta | 60.7 | — | Reasoning, Large, Text Generation, Open Weight |
| 15 | Alibaba | 60.0 | p84 | Code Assistant, Medium, Reasoning, Text Generation, Open Weight |
| 16 | Anthropic | 59.2 | — | Multimodal, Small, Text Generation, Reasoning, Proprietary, Large |
| 17 | OpenAI | 58.5 | p82 | Code Assistant, Small, Text Generation, Multimodal, Proprietary |
| 18 | 58.0 | p82 | Reasoning, Medium, Code Assistant, Text Generation, Open Weight |
| 19 | OpenAI | 57.0 | — | Code Assistant, Small, Text Generation, Multimodal, Proprietary |
| 20 | Alibaba | 57.0 | p81 | Code Assistant, Small, Text Generation, Reasoning, Open Weight, Medium |
| 21 | DeepSeek | 56.2 | — | Text Generation, Small, Code Assistant, Open Weight |
| 22 | Mistral AI | 55.2 | — | Text Generation, Small, Reasoning, Proprietary |
| 23 | 54.7 | — | Multimodal, Small, Text Generation, Reasoning, Proprietary |
| 24 | Anthropic | 54.7 | — | Code Assistant, Small, Text Generation, Multimodal, Reasoning, Proprietary |
| 25 | xAI | 54.3 | — | Multimodal, Small, Text Generation, Reasoning, Proprietary |
| 26 | Anthropic | 54.2 | p78 | Multimodal, Small, Text Generation, Proprietary |
| 27 | Microsoft | 52.0 | p75 | Multimodal, Medium, Text Generation, Open Source, Small |
