tinyanalysis

OCR models / ChatDOC

OCRFlux 3B

ChatDOC · released 2025-06-12 · data checked 2026-11-06

OCRFlux 3B is a 4.1B-parameter OCR and document-parsing model from ChatDOC (CN), released 2025-06-12 under the Not stated license. It has no published score on OmniDocBench v1.6 or olmOCR-Bench yet. It needs about 9.1 GB of memory at 16-bit.

Fine-tuned from Qwen2.5-VL 3B. Merges tables and paragraphs split across pages. Reports only its own OCRFlux benchmarks.

OmniDocBench v1.6
—
Not reported
olmOCR-Bench
—
Not reported
Parameters
4.1B
OCR specialist
Memory to run
~9.1 GB
16-bit weights plus about 1 GB

Benchmarks

No published score on these benchmarks yet. Know of one? Send it.

See how to read the ranking.

Specifications

Lab
ChatDOC (CN)
Type
OCR / document parsing specialist
Total parameters
4.1B
License
Not stated
No license is stated on the model card.
Memory to run
~9.1 GB at 16-bit
Released
2025-06-12
Weights and usage on Hugging Face ↗

Closest alternatives

  • OXR 1.0 (1.5B, 97.03 on OmniDocBench, OXR Open Source License)
  • Youtu-Parsing-Omni (5.3B, 96.96 on OmniDocBench, Youtu-Parsing license)
  • TeleOCR (1.4B, 96.91 on OmniDocBench, Apache-2.0)
  • Xiaomi-OCR-0 (870M, 96.83 on OmniDocBench, Apache-2.0)

Sources

Something wrong here? Report it.