deepseek-ai/DeepSeek-OCR

Note

This is a remotely-sourced model from the deepseek_ocr plugin, maintained by the community. It is not part of FiftyOne core and may have special installation requirements. Please review the plugin documentation and license before use.

DeepSeek-OCR is an open-source vision-language model (VLM) developed by DeepSeek to perform optical character recognition (OCR) and context compression for long and complex documents.

Details

• Model name: deepseek-ai/DeepSeek-OCR

• Model source: https://huggingface.co/deepseek-ai/DeepSeek-OCR

• Model author: DeepSeek

• Model license: MIT

• Exposes embeddings? no

• Tags: detection, ocr, VLM

Requirements

• Packages: huggingface-hub, transformers==4.46.3, tokenizers==0.20.3, torch, torchvision, addict, einops

• CPU support

  • yes

• GPU support

  • yes

Example usage

import fiftyone as fo
import fiftyone.zoo as foz

foz.register_zoo_model_source("https://github.com/harpreetsahota204/deepseek_ocr")

dataset = foz.load_zoo_dataset(
    "coco-2017",
    split="validation",
    dataset_name=fo.get_default_dataset_name(),
    max_samples=50,
    shuffle=True,
)

model = foz.load_zoo_model("deepseek-ai/DeepSeek-OCR")

dataset.apply_model(model, label_field="predictions")

session = fo.launch_app(dataset)