Real-world perception-centric reasoning on V* (test)
84.25AccuracyGLM-9B-DeltaThinker
Evaluation Results
| Method | Links | |
|---|---|---|
| GLM-9B-DeltaThinkerAlgorithm=DeltaPrompts2026.05 | 84.25 | |
| Qwen-8B-DeltaThinkerAlgorithm=DeltaPrompts2026.05 | 82.73 | |
| Vision-R1-7B2026.05 | 81.15 | |
| GLM-4.1V-9B-Thinking2026.05 | 79.58 | |
| Qwen3-VL-8B-Thinking2026.05 | 76.96 | |
| ARES-RL-7B2026.05 | 71.2 | |
| REVisual-R12026.05 | 69.11 | |
| Bee-8B-RL2026.05 | 58.12 |