All tasks / Models
Dolphin-v2
ByteDance's 3B open-weights document-parsing model for digital and photographed pages, built on Qwen2.5-VL-3B.
- ByteDance
- Open weights
- Released 1 Dec 2025
Where it ranks
- ADocuments & OCRModels · Local · 90 on OmniDocBench v1.6 (end-to-end, v1.6_full)
Facts
- Vendor
- ByteDance
- Family
- Dolphin
- Released
- 1 Dec 2025
- Weights
- Open: download and run it yourself
- License
- Qwen Research License (non-commercial; repository LICENSE)
- Parameters
- 3.75B
- GPU memory to run locally
- ~9 GB (Estimate, no official figure: BF16 weights ≈ 3.75 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 |
|---|---|---|---|---|
| OmniDocBench v1.6 (end-to-end, v1.6_full) | 90 | Dolphin-v2Overall score in the README's end-to-end table 'OmniDocBench (v1.6_full)' (last leaderboard update 2026-09-11). | 11 Sep 2026 | github.com |
Where you can use it
No tracked tool lists this model yet; as open weights it runs in local runtimes.
Earlier versions
- Dolphin-1.5 · —