tinyanalysis

OCR models / Tencent (Youtu)

Youtu-Parsing-Omni

Tencent (Youtu) · released 2026-10-09 · data checked 2026-11-06

Youtu-Parsing-Omni is a 5.3B-parameter OCR and document-parsing model from Tencent (Youtu) (CN), released 2026-10-09 under the Youtu-Parsing license license. It scores 96.96 on OmniDocBench v1.6 (#2 of 33). It needs about 11.7 GB of memory at 16-bit.

Parses documents and also charts, natural images, audio and video into one JSON schema.

OmniDocBench v1.6
96.96
#2 of 33 models
olmOCR-Bench
—
Not reported
Parameters
5.3B
OCR specialist
Memory to run
~11.7 GB
16-bit weights plus about 1 GB

Benchmarks

OmniDocBench v1.6 Lab-reported96.96
Best: OXR 1.0 97.03

The tick marks the best score on each benchmark. See how to read the ranking.

Specifications

Lab
Tencent (Youtu) (CN)
Type
OCR / document parsing specialist
Total parameters
5.3B
License
Youtu-Parsing license
Custom Tencent license; check its terms before commercial use.
Memory to run
~11.7 GB at 16-bit
Released
2026-10-09
Weights and usage on Hugging Face ↗

Closest alternatives

  • TeleOCR (1.4B, 96.91 on OmniDocBench, Apache-2.0)
  • OXR 1.0 (1.5B, 97.03 on OmniDocBench, OXR Open Source License)
  • Xiaomi-OCR-0 (870M, 96.83 on OmniDocBench, Apache-2.0)
  • OvisOCR3 (850M, 96.74 on OmniDocBench, Apache-2.0)

Sources

Something wrong here? Report it.