OCR models / Tencent (Hunyuan)

HunyuanOCR 1.0

Name: HunyuanOCR 1.0
Author: Tencent (Hunyuan)

Tencent (Hunyuan) · released 2025-11-25 · data checked 2026-11-06

HunyuanOCR 1.0 is a 1B-parameter OCR and document-parsing model from Tencent (Hunyuan) (CN), released 2025-11-25 under the Tencent Hunyuan Community License license. It scores 89.95 on OmniDocBench v1.6 (#27 of 33). It needs about 3 GB of memory at 16-bit.

Succeeded by HunyuanOCR 1.5. Its weights are in the repository's v1.0 folder.

OmniDocBench v1.6

89.95

#27 of 33 models

olmOCR-Bench

—

Not reported

Parameters

OCR specialist

Memory to run

~3 GB

16-bit weights plus about 1 GB

Benchmarks

OmniDocBench v1.6 Leaderboard89.95

Best: OXR 1.0 97.03

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

Specifications

Lab: Tencent (Hunyuan) (CN)
Type: OCR / document parsing specialist
Total parameters: 1B
License: Tencent Hunyuan Community License
Custom Tencent license; check its terms before commercial use.
Memory to run: ~3 GB at 16-bit
Released: 2025-11-25

Weights and usage on Hugging Face ↗

Closest alternatives

DeepSeek-OCR 2 (3.4B, 90.25 on OmniDocBench, Apache-2.0)
Dolphin v2 (3.8B, 89.50 on OmniDocBench, Not stated)
dots.ocr (3B, 90.77 on OmniDocBench, MIT)
jina-ocr-v1 (3.4B, 91.14 on OmniDocBench, CC-BY-NC-4.0)

Sources

Something wrong here? Report it.