---
title: DeepSeek-OCR-GGUF-CUDA
canonical_url: "https://www.modelscope.ai/models/NexaAI/DeepSeek-OCR-GGUF-CUDA"
md_url: "https://www.modelscope.ai/models/NexaAI/DeepSeek-OCR-GGUF-CUDA.md"
repository: NexaAI/DeepSeek-OCR-GGUF-CUDA
last_updated: 2025-11-21
base_model:
  - deepseek-ai/DeepSeek-OCR
base_model_relation: quantized
library_name:
  - gguf
  - pytorch
frameworks:
  - pytorch
language:
  - multilingual
downloads: 127
stars: 0
tags:
  - deepseek
  - vision-language
  - ocr
  - document-parse
  - gguf
---

# DeepSeek-OCR-GGUF-CUDA

> DeepSeek-OCR-GGUF-CUDA - An open-source model by NexaAI on ModelScope. [!NOTE] Note currently only NexaSDK supports this model's GGUF.

NexaAI/DeepSeek-OCR-GGUF-CUDA is a machine learning model on ModelScope. derived from deepseek-ai/DeepSeek-OCR.

- **Repository**: NexaAI/DeepSeek-OCR-GGUF-CUDA
- **Base model**: deepseek-ai/DeepSeek-OCR
- **Tags**: deepseek, vision-language, ocr, document-parse, gguf
- **Downloads**: 127
- **Stars**: 0
- **Last updated**: 2025-11-21

Source: https://www.modelscope.ai/models/NexaAI/DeepSeek-OCR-GGUF-CUDA

---

# DeepSeek OCR

> [!NOTE]  
> Note currently only [NexaSDK](https://github.com/NexaAI/nexa-sdk) supports this model's GGUF.

## Quickstart

1. **Install [NexaSDK](https://github.com/NexaAI/nexa-sdk)**
2. Run the model locally with one line of code:

```bash
nexa infer NexaAI/DeepSeek-OCR-GGUF
 ```
3. Then drag your image to terminal or type into the image path


case 1 : extract text
```bash
<your-image-path> Free OCR.
```
case 2 : extract bounding box
```bash
<your-image-path> <|grounding|>Convert the document to markdown. 
```

## Model Description
**DeepSeek OCR** is a high-accuracy optical character recognition model built for extracting text from complex visual inputs such as documents, screenshots, receipts, and natural scenes.  
It combines vision-language modeling with efficient visual encoders to achieve superior recognition of multi-language and multi-layout text while remaining lightweight enough for edge or on-device deployment.

## Features
- **Multilingual OCR** — recognizes printed and handwritten text across major global languages.  
- **Document Layout Understanding** — preserves structure such as tables, paragraphs, and titles.  
- **Scene Text Recognition** — robust against lighting, distortion, and low-quality captures.  
- **Lightweight & Fast** — optimized for CPU and GPU acceleration.  
- **End-to-End Pipeline** — supports image-to-text and structured JSON output.  

## Use Cases
- Digitizing scanned documents or PDFs  
- Extracting text from mobile camera inputs or screenshots  
- Invoice and receipt parsing  
- OCR-based search and indexing systems  
- Visual question answering or document agents  

## Inputs and Outputs
**Input:**  
- Image file (JPEG, PNG, or tensor array)  
- Optional parameters for language hints or layout detection  

**Output:**  
- Extracted text (plain text or structured format with bounding boxes)  
- Confidence scores per word or region  

## Integration
DeepSeek OCR can be integrated through:  
- Python API (`pip install deepseek-ocr`)  
- REST or gRPC endpoints for server deployment  

## License
This model is released under the **Apache 2.0 License**, allowing commercial use, modification, and redistribution with attribution.