From d033c1cba23b9b3b864ad233a8886b24f6f69aa6 Mon Sep 17 00:00:00 2001 From: "hongliang.yuan" Date: Fri, 24 Oct 2025 13:58:54 +0800 Subject: [PATCH] add deepseek ocr --- .../deepseek-ocr/transformers/README.md | 41 +++++++++++++++++++ 1 file changed, 41 insertions(+) create mode 100644 models/multimodal/vision_language_model/deepseek-ocr/transformers/README.md diff --git a/models/multimodal/vision_language_model/deepseek-ocr/transformers/README.md b/models/multimodal/vision_language_model/deepseek-ocr/transformers/README.md new file mode 100644 index 00000000..d7cc0f29 --- /dev/null +++ b/models/multimodal/vision_language_model/deepseek-ocr/transformers/README.md @@ -0,0 +1,41 @@ +# DeepSeek-OCR (Transformers) + +## Model Description + +DeepSeek-OCR is DeepSeek's optical character recognition (OCR) system designed to extract text from images and documents. Here are its key features: + +- Text Detection: Identifies text regions within images, PDFs, and other document formats +- Text Recognition: Converts detected text areas into machine-readable text +- Multi-format Support: Works with various file types including images (JPG, PNG, etc.) and PDF documents +- High Accuracy: Provides precise text extraction with advanced AI models + +## Supported Environments + +| GPU | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release | +| :----: | :----: | :----: | +| MR-V100 | 4.3.0 | 25.12 | + +## Model Preparation + +### Prepare Resources + +- Model: + +### Install Dependencies + +In order to run the model smoothly, you need to get the sdk from [resource center](https://support.iluvatar.com/#/ProductLine?id=2) of Iluvatar CoreX official website. + +```bash +pip install transformers==4.46.3 einops easydict addict matplotlib +``` + +## Model Inference + +```bash +git clone https://github.com/deepseek-ai/DeepSeek-OCR.git +cd DeepSeek-OCR/DeepSeek-OCR-master/DeepSeek-OCR-hf/ +# Change image_file and output_path value with your own config +python3 run_dpsk_ocr.py +``` + +## Model Results \ No newline at end of file -- Gitee