Science Question Answering on GPQA (Score Metric)
61.5ScoreQwen 3 VL 8B Think
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen 3 VL 8B Think2025.12 | 61.5 | |
| Qwen 3 8B2025.12 | 57.7 | |
| OR Nemotron 7B2025.12 | 56.6 | |
| Nemotron Nano 9B v22025.12 | 56.2 | |
| DS-R1 Qwen 7B2025.12 | 54.4 | |
| Olmo 3 7B ThinkStage=DPO2025.12 | 48.6 | |
| OpenThinker3 7B2025.12 | 47.6 | |
| Olmo 3 7B ThinkStage=Final Think2025.12 | 46.2 | |
| Olmo 3 7B ThinkStage=SFT2025.12 | 45.8 |