Instructions to use mingyi456/DeepSeek-OCR-DF11 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mingyi456/DeepSeek-OCR-DF11 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="mingyi456/DeepSeek-OCR-DF11")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("mingyi456/DeepSeek-OCR-DF11", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use mingyi456/DeepSeek-OCR-DF11 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "mingyi456/DeepSeek-OCR-DF11" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mingyi456/DeepSeek-OCR-DF11", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/mingyi456/DeepSeek-OCR-DF11
- SGLang
How to use mingyi456/DeepSeek-OCR-DF11 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "mingyi456/DeepSeek-OCR-DF11" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mingyi456/DeepSeek-OCR-DF11", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "mingyi456/DeepSeek-OCR-DF11" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mingyi456/DeepSeek-OCR-DF11", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use mingyi456/DeepSeek-OCR-DF11 with Docker Model Runner:
docker model run hf.co/mingyi456/DeepSeek-OCR-DF11
| { | |
| "dfloat11_config": { | |
| "version": "0.5.0", | |
| "threads_per_block": [ | |
| 512 | |
| ], | |
| "bytes_per_thread": 8, | |
| "pattern_dict": { | |
| "lm_head": [], | |
| "model\\.embed_tokens": [], | |
| "model\\.layers\\.0": [ | |
| "self_attn.q_proj", | |
| "self_attn.k_proj", | |
| "self_attn.v_proj", | |
| "self_attn.o_proj", | |
| "mlp.gate_proj", | |
| "mlp.up_proj", | |
| "mlp.down_proj" | |
| ], | |
| "model\\.layers\\.[1-9]\\d*": [ | |
| "self_attn.q_proj", | |
| "self_attn.k_proj", | |
| "self_attn.v_proj", | |
| "self_attn.o_proj", | |
| "mlp.experts.0.gate_proj", | |
| "mlp.experts.0.up_proj", | |
| "mlp.experts.0.down_proj", | |
| "mlp.experts.1.gate_proj", | |
| "mlp.experts.1.up_proj", | |
| "mlp.experts.1.down_proj", | |
| "mlp.experts.2.gate_proj", | |
| "mlp.experts.2.up_proj", | |
| "mlp.experts.2.down_proj", | |
| "mlp.experts.3.gate_proj", | |
| "mlp.experts.3.up_proj", | |
| "mlp.experts.3.down_proj", | |
| "mlp.experts.4.gate_proj", | |
| "mlp.experts.4.up_proj", | |
| "mlp.experts.4.down_proj", | |
| "mlp.experts.5.gate_proj", | |
| "mlp.experts.5.up_proj", | |
| "mlp.experts.5.down_proj", | |
| "mlp.experts.6.gate_proj", | |
| "mlp.experts.6.up_proj", | |
| "mlp.experts.6.down_proj", | |
| "mlp.experts.7.gate_proj", | |
| "mlp.experts.7.up_proj", | |
| "mlp.experts.7.down_proj", | |
| "mlp.experts.8.gate_proj", | |
| "mlp.experts.8.up_proj", | |
| "mlp.experts.8.down_proj", | |
| "mlp.experts.9.gate_proj", | |
| "mlp.experts.9.up_proj", | |
| "mlp.experts.9.down_proj", | |
| "mlp.experts.10.gate_proj", | |
| "mlp.experts.10.up_proj", | |
| "mlp.experts.10.down_proj", | |
| "mlp.experts.11.gate_proj", | |
| "mlp.experts.11.up_proj", | |
| "mlp.experts.11.down_proj", | |
| "mlp.experts.12.gate_proj", | |
| "mlp.experts.12.up_proj", | |
| "mlp.experts.12.down_proj", | |
| "mlp.experts.13.gate_proj", | |
| "mlp.experts.13.up_proj", | |
| "mlp.experts.13.down_proj", | |
| "mlp.experts.14.gate_proj", | |
| "mlp.experts.14.up_proj", | |
| "mlp.experts.14.down_proj", | |
| "mlp.experts.15.gate_proj", | |
| "mlp.experts.15.up_proj", | |
| "mlp.experts.15.down_proj", | |
| "mlp.experts.16.gate_proj", | |
| "mlp.experts.16.up_proj", | |
| "mlp.experts.16.down_proj", | |
| "mlp.experts.17.gate_proj", | |
| "mlp.experts.17.up_proj", | |
| "mlp.experts.17.down_proj", | |
| "mlp.experts.18.gate_proj", | |
| "mlp.experts.18.up_proj", | |
| "mlp.experts.18.down_proj", | |
| "mlp.experts.19.gate_proj", | |
| "mlp.experts.19.up_proj", | |
| "mlp.experts.19.down_proj", | |
| "mlp.experts.20.gate_proj", | |
| "mlp.experts.20.up_proj", | |
| "mlp.experts.20.down_proj", | |
| "mlp.experts.21.gate_proj", | |
| "mlp.experts.21.up_proj", | |
| "mlp.experts.21.down_proj", | |
| "mlp.experts.22.gate_proj", | |
| "mlp.experts.22.up_proj", | |
| "mlp.experts.22.down_proj", | |
| "mlp.experts.23.gate_proj", | |
| "mlp.experts.23.up_proj", | |
| "mlp.experts.23.down_proj", | |
| "mlp.experts.24.gate_proj", | |
| "mlp.experts.24.up_proj", | |
| "mlp.experts.24.down_proj", | |
| "mlp.experts.25.gate_proj", | |
| "mlp.experts.25.up_proj", | |
| "mlp.experts.25.down_proj", | |
| "mlp.experts.26.gate_proj", | |
| "mlp.experts.26.up_proj", | |
| "mlp.experts.26.down_proj", | |
| "mlp.experts.27.gate_proj", | |
| "mlp.experts.27.up_proj", | |
| "mlp.experts.27.down_proj", | |
| "mlp.experts.28.gate_proj", | |
| "mlp.experts.28.up_proj", | |
| "mlp.experts.28.down_proj", | |
| "mlp.experts.29.gate_proj", | |
| "mlp.experts.29.up_proj", | |
| "mlp.experts.29.down_proj", | |
| "mlp.experts.30.gate_proj", | |
| "mlp.experts.30.up_proj", | |
| "mlp.experts.30.down_proj", | |
| "mlp.experts.31.gate_proj", | |
| "mlp.experts.31.up_proj", | |
| "mlp.experts.31.down_proj", | |
| "mlp.experts.32.gate_proj", | |
| "mlp.experts.32.up_proj", | |
| "mlp.experts.32.down_proj", | |
| "mlp.experts.33.gate_proj", | |
| "mlp.experts.33.up_proj", | |
| "mlp.experts.33.down_proj", | |
| "mlp.experts.34.gate_proj", | |
| "mlp.experts.34.up_proj", | |
| "mlp.experts.34.down_proj", | |
| "mlp.experts.35.gate_proj", | |
| "mlp.experts.35.up_proj", | |
| "mlp.experts.35.down_proj", | |
| "mlp.experts.36.gate_proj", | |
| "mlp.experts.36.up_proj", | |
| "mlp.experts.36.down_proj", | |
| "mlp.experts.37.gate_proj", | |
| "mlp.experts.37.up_proj", | |
| "mlp.experts.37.down_proj", | |
| "mlp.experts.38.gate_proj", | |
| "mlp.experts.38.up_proj", | |
| "mlp.experts.38.down_proj", | |
| "mlp.experts.39.gate_proj", | |
| "mlp.experts.39.up_proj", | |
| "mlp.experts.39.down_proj", | |
| "mlp.experts.40.gate_proj", | |
| "mlp.experts.40.up_proj", | |
| "mlp.experts.40.down_proj", | |
| "mlp.experts.41.gate_proj", | |
| "mlp.experts.41.up_proj", | |
| "mlp.experts.41.down_proj", | |
| "mlp.experts.42.gate_proj", | |
| "mlp.experts.42.up_proj", | |
| "mlp.experts.42.down_proj", | |
| "mlp.experts.43.gate_proj", | |
| "mlp.experts.43.up_proj", | |
| "mlp.experts.43.down_proj", | |
| "mlp.experts.44.gate_proj", | |
| "mlp.experts.44.up_proj", | |
| "mlp.experts.44.down_proj", | |
| "mlp.experts.45.gate_proj", | |
| "mlp.experts.45.up_proj", | |
| "mlp.experts.45.down_proj", | |
| "mlp.experts.46.gate_proj", | |
| "mlp.experts.46.up_proj", | |
| "mlp.experts.46.down_proj", | |
| "mlp.experts.47.gate_proj", | |
| "mlp.experts.47.up_proj", | |
| "mlp.experts.47.down_proj", | |
| "mlp.experts.48.gate_proj", | |
| "mlp.experts.48.up_proj", | |
| "mlp.experts.48.down_proj", | |
| "mlp.experts.49.gate_proj", | |
| "mlp.experts.49.up_proj", | |
| "mlp.experts.49.down_proj", | |
| "mlp.experts.50.gate_proj", | |
| "mlp.experts.50.up_proj", | |
| "mlp.experts.50.down_proj", | |
| "mlp.experts.51.gate_proj", | |
| "mlp.experts.51.up_proj", | |
| "mlp.experts.51.down_proj", | |
| "mlp.experts.52.gate_proj", | |
| "mlp.experts.52.up_proj", | |
| "mlp.experts.52.down_proj", | |
| "mlp.experts.53.gate_proj", | |
| "mlp.experts.53.up_proj", | |
| "mlp.experts.53.down_proj", | |
| "mlp.experts.54.gate_proj", | |
| "mlp.experts.54.up_proj", | |
| "mlp.experts.54.down_proj", | |
| "mlp.experts.55.gate_proj", | |
| "mlp.experts.55.up_proj", | |
| "mlp.experts.55.down_proj", | |
| "mlp.experts.56.gate_proj", | |
| "mlp.experts.56.up_proj", | |
| "mlp.experts.56.down_proj", | |
| "mlp.experts.57.gate_proj", | |
| "mlp.experts.57.up_proj", | |
| "mlp.experts.57.down_proj", | |
| "mlp.experts.58.gate_proj", | |
| "mlp.experts.58.up_proj", | |
| "mlp.experts.58.down_proj", | |
| "mlp.experts.59.gate_proj", | |
| "mlp.experts.59.up_proj", | |
| "mlp.experts.59.down_proj", | |
| "mlp.experts.60.gate_proj", | |
| "mlp.experts.60.up_proj", | |
| "mlp.experts.60.down_proj", | |
| "mlp.experts.61.gate_proj", | |
| "mlp.experts.61.up_proj", | |
| "mlp.experts.61.down_proj", | |
| "mlp.experts.62.gate_proj", | |
| "mlp.experts.62.up_proj", | |
| "mlp.experts.62.down_proj", | |
| "mlp.experts.63.gate_proj", | |
| "mlp.experts.63.up_proj", | |
| "mlp.experts.63.down_proj", | |
| "mlp.shared_experts.gate_proj", | |
| "mlp.shared_experts.up_proj", | |
| "mlp.shared_experts.down_proj" | |
| ], | |
| "model\\.sam_model\\.blocks\\.\\d+": [ | |
| "attn.qkv", | |
| "attn.proj", | |
| "mlp.lin1", | |
| "mlp.lin2" | |
| ], | |
| "model\\.vision_model\\.embeddings\\.position_embedding": [], | |
| "model\\.vision_model\\.transformer\\.layers\\.\\d+": [ | |
| "self_attn.qkv_proj", | |
| "self_attn.out_proj", | |
| "mlp.fc1", | |
| "mlp.fc2" | |
| ], | |
| "model\\.projector\\.layers": [] | |
| } | |
| } | |
| } |