Visual Reasoning on VLMsAreBlind
61.9VLMsAreBlind ScoreQwen3-VL
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3-VLParams=8B2026.04 | 61.9 | |
| InternVL3.5Params=8B2026.04 | 45.4 | |
| InternVL3Params=8B2026.04 | 42 | |
| Qwen2.5-VLParams=7B2026.04 | 40.3 | |
| GPT-5 Nano2026.04 | 40.2 | |
| UniDriveVLAParams=8B2026.04 | 26.6 |