Real-world perception-centric reasoning on Real-world perception-centric reasoning suite (test)
55.53Average ScoreGLM-9B-DeltaThinker
Evaluation results
| Method | Links | |
|---|---|---|
| GLM-9B-DeltaThinkerAlgorithm=DeltaPrompts2026.05 | 55.53 | |
| Qwen-8B-DeltaThinkerAlgorithm=DeltaPrompts2026.05 | 54.12 | |
| GLM-4.1V-9B-Thinking2026.05 | 49.75 | |
| Vision-R1-7B2026.05 | 48.67 | |
| REVisual-R12026.05 | 48.06 | |
| Qwen3-VL-8B-Thinking2026.05 | 46.92 | |
| ARES-RL-7B2026.05 | 46.45 | |
| Bee-8B-RL2026.05 | 40.43 |