Image-Text-to-Text
Transformers
Safetensors
multilingual
deepseekocr
feature-extraction
deepseek
vision-language
ocr
custom_code
Instructions to use specsGuy/Deepseek-ocr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use specsGuy/Deepseek-ocr with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="specsGuy/Deepseek-ocr", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("specsGuy/Deepseek-ocr", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use specsGuy/Deepseek-ocr with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "specsGuy/Deepseek-ocr" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "specsGuy/Deepseek-ocr", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/specsGuy/Deepseek-ocr
- SGLang
How to use specsGuy/Deepseek-ocr 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 "specsGuy/Deepseek-ocr" \ --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": "specsGuy/Deepseek-ocr", "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 "specsGuy/Deepseek-ocr" \ --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": "specsGuy/Deepseek-ocr", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use specsGuy/Deepseek-ocr with Docker Model Runner:
docker model run hf.co/specsGuy/Deepseek-ocr
Update modeling_deepseekocr.py
Browse files- modeling_deepseekocr.py +1 -1
modeling_deepseekocr.py
CHANGED
|
@@ -620,7 +620,7 @@ class DeepseekOCRForCausalLM(DeepseekV2ForCausalLM):
|
|
| 620 |
if past_key_values is not None:
|
| 621 |
if isinstance(past_key_values, Cache):
|
| 622 |
cache_length = past_key_values.get_seq_length()
|
| 623 |
-
past_length = past_key_values
|
| 624 |
max_cache_length = past_key_values.get_max_length()
|
| 625 |
else:
|
| 626 |
cache_length = past_length = past_key_values[0][0].shape[2]
|
|
|
|
| 620 |
if past_key_values is not None:
|
| 621 |
if isinstance(past_key_values, Cache):
|
| 622 |
cache_length = past_key_values.get_seq_length()
|
| 623 |
+
past_length = getattr(past_key_values, "seen_tokens", cache_length)
|
| 624 |
max_cache_length = past_key_values.get_max_length()
|
| 625 |
else:
|
| 626 |
cache_length = past_length = past_key_values[0][0].shape[2]
|