All tasks / Models
MinerU2.5-Pro-2605-1.2B
OpenDataLab's open-weights 1.2B vision-language model for document parsing, used by the MinerU toolkit.
- OpenDataLab (Shanghai AI Laboratory)
- Open weights
- Released 21 May 2026
Where it ranks
- SDocuments & OCRModels · Local · 96 on OmniDocBench v1.6 (end-to-end, v1.6_full)
Facts
- Vendor
- OpenDataLab (Shanghai AI Laboratory)
- Family
- MinerU (VLM)
- Released
- 21 May 2026
- Weights
- Open: download and run it yourself
- License
- Apache-2.0
- Parameters
- 1.16B
- GPU memory to run locally
- ~4 GB (Estimate, no official figure: BF16 weights ≈ 1.16 B × 2 GB + ~1.5 GB runtime overhead.)
- Access
- Free (open weights)
- Last verified
- 18 Oct 2026
Price from huggingface.co, 18 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| OmniDocBench v1.6 (end-to-end, v1.6_full) | 96 | MinerU2.5-ProOverall score in the README's end-to-end table 'OmniDocBench (v1.6_full)' (last leaderboard update 2026-09-11); the README's model-information table names the evaluated weights MinerU2.5-Pro-2605-1.2B. | 11 Sep 2026 | github.com |
| ParseBench (models) | 72.8% | MinerU2.5-Pro-2605-1.2BOverall (mean of tables, charts, content faithfulness, semantic formatting, visual grounding) from leaderboard.csv in the ParseBench repo (last leaderboard commit 2026-10-09). | 9 Oct 2026 | github.com |
Where you can use it
- SMinerUParsers & APIs · Local
Earlier versions
- MinerU2.5-Pro-2604-1.2B · 2 Apr 2026
- MinerU2.5-2509-1.2B · 17 Sep 2025
- MinerU2.0-2505-0.9B · May 2025