What are the hardware requirements for HunyuanOCR?

HunyuanOCR requires approximately 2.2 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed HunyuanOCR and how large is it?

HunyuanOCR is developed by tencent, a model with 1.1b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate HunyuanOCR?

HunyuanOCR is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

HunyuanOCR by tencent — VRAM 2.2GB, 128K context

Name: HunyuanOCR
Author: tencent

HunyuanOCR is an open-source language model by tencent. Features: 1.1b LLM, VRAM: 2.2GB, Context: 128K, License: other, LLM Explorer Score: 0.36.

Arxiv:2607.04884 Conversational Document-parsing En Endpoints compatible Eval-results Hunyuan vl Image-text-to-text Information-extraction Multilingual Ocr Region:us Safetensors Text-image-translation Text-spotting Vision Vision-language-model Zh

Model Card on HF 🤗: https://huggingface.co/tencent/HunyuanOCR

HunyuanOCR Parameters and Internals

LLM Name	HunyuanOCR
Repository 🤗	https://huggingface.co/tencent/HunyuanOCR
Model Size	1.1b
Required VRAM	2.2 GB
Updated	2026-08-09
Maintainer	tencent
Model Type	hunyuan_vl
Model Files	2.2 GB
Supported Languages	en zh
Model Architecture	HunYuanVLForConditionalGeneration
License	other
Context Length	131072
Model Max Length	131072
Transformers Version	4.57.1
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<｜hy_▁pad▁｜>
Vocabulary Size	120818

Rank the HunyuanOCR capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation