Visual Reasoning on TeT ColorBlind
29.33Pass@1DDP
Evaluation results
| Method | Links | |
|---|---|---|
| DDPbackbone=Gemini-3-Pro2026.04 | 29.33 | |
| Gemini-3-Pro2026.04 | 15.33 | |
| OpenAI o12026.04 | 0 | |
| Claude-3-Sonnet2026.04 | 0 | |
| Gemini-2.5-Pro2026.04 | 0 | |
| Qwen2.5-VL-72B2026.04 | 0 | |
| InternVL3-78B2026.04 | 0 | |
| MiniCPM-V-2.62026.04 | 0 |