All tasks / Models
Surya OCR 2
Datalab's open-weights small OCR model for text, layout, reading order and tables in 90+ languages; the engine behind Marker.
- Datalab
- Open weights
- Released 14 May 2026
Where it ranks
- –Documents & OCRModels · Local · no result on this task’s benchmarks yet
Facts
- Vendor
- Datalab
- Family
- Surya OCR
- Released
- 14 May 2026
- Weights
- Open: download and run it yourself
- License
- OpenRAIL (Hugging Face license tag); surya code Apache-2.0
- Parameters
- 0.69B
- GPU memory to run locally
- ~3 GB (Estimate, no official figure: BF16 weights ≈ 0.69 B × 2 GB + ~1.5 GB runtime overhead. Size is the Hugging Face safetensors total.)
- Access
- Free (open weights)
- Last verified
- 19 Oct 2026
Price from huggingface.co, 19 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| ParseBench (models) | 64.8% | Surya OCR 2Overall (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
- –MarkerParsers & APIs · Local