Official benchmarks map / MDPBench
FINAL-Bench/Darwin-180B-RSI is ranked first on the MDPBench leaderboard with 83.65, 0.1% ahead of moonshotai/Kimi-K3. Read live from the Hugging Face API on November 6, 2026.
| # | Model | Organization | Score |
|---|---|---|---|
| 1 | FINAL-Bench/Darwin-180B-RSI | FINAL-Bench | 83.65 |
| 2 | moonshotai/Kimi-K3 | moonshotai | 83.6 |
| 3 | dots-studio/dots.mocr | dots-studio | 80.5 |
| 4 | datalab-to/chandra-ocr-2 | datalab-to | 79.7 |
| 5 | PaddlePaddle/PaddleOCR-VL-1.6 | PaddlePaddle | 78.9 |
| 6 | PaddlePaddle/PaddleOCR-VL-1.5 | PaddlePaddle | 78.3 |
| 7 | moonshotai/Kimi-K2.5 | moonshotai | 77.5 |
| 8 | dots-studio/dots.ocr | dots-studio | 76.5 |
| 9 | opendatalab/MinerU2.5-Pro-2605-1.2B | opendatalab | 71 |
| 10 | Qwen/Qwen3-VL-8B-Instruct | Qwen | 68.3 |
MDPBench is a dataset that Hugging Face marks as an official benchmark. Its leaderboard is built automatically from the .eval_results files that model repositories publish, so entries are mostly self-reported by model authors and evaluation settings differ between submissions. A verified tag appears only where Hugging Face marks an entry as verified.
Original board: huggingface.co/datasets/Delores-Lin/MDPBench
Part of the Hugging Face official benchmarks map by QuantID. Unofficial. Not affiliated with or endorsed by Hugging Face.