Question Answering on GPQA (val)
84GPQA ScoreGrok 3 Mini
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Grok 3 Mini2025.09 | 84 | 7.37 | 3 | |
| Qwen3 235B A22B Instruct 25072025.09 | 77.5 | 7.56 | 2 | |
| Kimi K2 09052025.09 | 75.8 | 6.81 | 8 | |
| Qwen3 Next 80B A3B Instruct2025.09 | 72.9 | 7.58 | 1 | |
| Llama 4 Maverick2025.09 | 69.8 | 7.35 | 4 | |
| DeepSeek V3 03242025.09 | 68.4 | 7.06 | 6 | |
| Gemini 2.5 Flash Lite2025.09 | 64.6 | 6.79 | 9 | |
| Gemini 2.0 Flash2025.09 | 62.1 | 7.17 | 5 | |
| Llama 4 Scout2025.09 | 57.2 | 6.86 | 7 | |
| Gemini 2.0 Flash Lite2025.09 | 51.5 | 6.51 | 11 | |
| Llama 3.3 70b Instruct2025.09 | 50.5 | 6.6 | 10 | |
| Qwen2.5 72B Instruct2025.09 | 49 | 6.27 | 12 | |
| Llama 3.1 70B Instruct2025.09 | 41.7 | 5.95 | 13 | |
| GPT 4o Mini2025.09 | 40.2 | 5.31 | 14 | |
| GPT 3.5 Turbo2025.09 | 30.8 | 4.81 | 15 |