---
title: Unlimited-OCR-2-8bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/Unlimited-OCR-2-8bit"
md_url: "https://www.modelscope.cn/models/mlx-community/Unlimited-OCR-2-8bit.md"
repository: mlx-community/Unlimited-OCR-2-8bit
last_updated: 2026-10-27
license: mit
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
model_type:
  - unlimited-ocr
architectures:
  - UnlimitedOCRForCausalLM
base_model:
  - baidu/Unlimited-OCR-2
base_model_relation: quantized
parameters: 1.2B
tensor_type:
  - BF16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - multilingual
downloads: 61
stars: 0
tags:
  - mlx
  - ocr
  - vision-language
---

# Unlimited-OCR-2-8bit

> Unlimited-OCR-2-8bit - mlx-community 在 ModelScope 开源的模型。mlx-community/Unlimited-OCR-2-8bit

mlx-community/Unlimited-OCR-2-8bit 是 ModelScope 魔搭社区上的 1.2B 参数image-text-to-text模型，采用 mit 许可，基于 baidu/Unlimited-OCR-2 构建。

- **Repository**: mlx-community/Unlimited-OCR-2-8bit
- **License**: mit
- **Tasks**: image-text-to-text
- **Parameters**: 1.2B
- **Base model**: baidu/Unlimited-OCR-2
- **Tags**: mlx, ocr, vision-language
- **Downloads**: 61
- **Stars**: 0
- **Last updated**: 2026-10-27

Source: https://www.modelscope.cn/models/mlx-community/Unlimited-OCR-2-8bit

---

# mlx-community/Unlimited-OCR-2-8bit

[`baidu/Unlimited-OCR-2`](https://huggingface.co/baidu/Unlimited-OCR-2) converted to MLX with
[mlx-vlm](https://github.com/Blaizzy/mlx-vlm): 8-bit affine quantization (group size 64) of the language model. The SAM and CLIP vision towers and the
per-layer DASH router stay in bfloat16.

Refer to the [original model card](https://huggingface.co/baidu/Unlimited-OCR-2) for details on the model.

## Use with mlx-vlm

Unlimited-OCR-2 support (page2d RoPE and DASH block-sparse attention) is being upstreamed to
mlx-vlm. Until it is released, install from the branch that adds it:

```bash
pip install -U "git+https://github.com/shuuul/mlx-vlm.git"
```

```bash
# One page (gundam mode: 1024 global view + up to 16 local crops)
mlx_vlm.generate --model mlx-community/Unlimited-OCR-2-8bit --image page.png \
    --prompt "Multi page parsing." --max-tokens 8192 --temperature 0

# Several pages in one request (one 1024 base view per page)
mlx_vlm.generate --model mlx-community/Unlimited-OCR-2-8bit --image p1.png p2.png \
    --prompt "Multi page parsing." --max-tokens 16384 --temperature 0
```

The output uses `<PAGE>`, `<BLOCK>label[x0, y0, x1, y1]<FORK>text` and `cross_*` / `append_*`
labels for blocks that continue across a page break; the upstream repository ships a converter
to Markdown.