Complex Reasoning on cvbench
54.9AccuracyQwen3-VL-4B + SynRL
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3-VL-4B + SynRLModel Scale=4B, SynRL Training=true2026.03 | 54.9 | |
| Qwen3-VL-8B + SynRLModel Scale=8B, SynRL Training=true2026.03 | 54.9 | |
| Qwen3-VL-4BModel Scale=4B, SynRL Training=false2026.03 | 54.1 | |
| Qwen3-VL-8BModel Scale=8B, SynRL Training=false2026.03 | 53.8 |