Unified Multimodal Parsing on OmniParsingBench Natural Image
69.72Overall ScoreGemini-3-Pro
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini-3-Pro2026.03 | 69.72 | 63.15 | 76.29 | |
| Logics-Parsing-Omni2026.03 | 62.46 | 50.53 | 74.38 | |
| Qwen3-VL-235B-A22BParameters=235B-A22B2026.03 | 59.34 | 50.62 | 68.05 | |
| Qwen3.5-397B-A17BParameters=397B-A17B2026.03 | 54.8 | 47.12 | 62.48 | |
| Qwen3-VL-30B-A3BParameters=30B-A3B2026.03 | 53.73 | 46.47 | 60.99 | |
| GPT-5.22026.03 | 52.84 | 46.07 | 59.61 | |
| Qwen3-Omni-30B-A3BParameters=30B-A3B2026.03 | 51.3 | 44.88 | 57.72 |