Question Answering on GPQA
51.1ScoreQwen 3 VL 8B Inst
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen 3 VL 8B Inststage=Instruct2025.12 | 51.1 | |
| Qwen 3 8Bstage=Instruct2025.12 | 44.6 | |
| Olmo 3 7B Instructstage=Final Instruct2025.12 | 40.4 | |
| Olmo 3 7B Instructstage=DPO2025.12 | 37.9 | |
| Qwen 2.5 7Bstage=Instruct2025.12 | 35.6 | |
| Granite 3.3 8B Inststage=Instruct2025.12 | 33 | |
| Qwen2-7BParameters=7B2024.07 | 31.8 | |
| OLMo 2 7B Inststage=Instruct2025.12 | 31.3 | |
| Olmo 3 7B Instructstage=SFT2025.12 | 30 | |
| Apertus 8B Inststage=Instruct2025.12 | 28.8 | |
| Qwen1.5-7BParameters=7B2024.07 | 26.7 | |
| Llama-3-8BParameters=8B2024.07 | 25.8 | |
| Gemma-7BParameters=7B2024.07 | 25.7 | |
| Mistral-7BParameters=7B2024.07 | 24.7 |