Instructions to use Deepdive404/Deepseek-fork with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Deepdive404/Deepseek-fork with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Deepdive404/Deepseek-fork", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Deepdive404/Deepseek-fork", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("Deepdive404/Deepseek-fork", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Deepdive404/Deepseek-fork with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Deepdive404/Deepseek-fork"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Deepdive404/Deepseek-fork",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Deepdive404/Deepseek-fork

SGLang

How to use Deepdive404/Deepseek-fork 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 "Deepdive404/Deepseek-fork" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Deepdive404/Deepseek-fork",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Deepdive404/Deepseek-fork" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Deepdive404/Deepseek-fork",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Deepdive404/Deepseek-fork with Docker Model Runner:
```
docker model run hf.co/Deepdive404/Deepseek-fork
```

Thank you for this fork, all of you.

by Bluky-BlueSky - opened Feb 5, 2025

Discussion

Bluky-BlueSky

Feb 5, 2025

But what I don't understand is that it has been several days, why not many people have found this fork or made similar forks.
（Maybe I asked a stupid question, because I don't know much about the operation mode of HuggingFace. I just Google it a little. English is not my mother tongue, so I may have made some mistakes.）
I just have some strange curiosity about the existence of this repository in this situation, especially the information of one contributor is "Deepdive404" and the information of another contributor is Mr. Ma.

Deepdive404

Owner Feb 5, 2025

Hello @Bluky-BlueSky

Huggingface does not have a fork feature per say, what we can do is duplicate a repo via one of their hosted apps (spaces) which is what I did.

Repo duplicator https://huggingface.co/spaces/huggingface-projects/repo_duplicator

Hope this helps

Bluky-BlueSky

Feb 6, 2025

Hi @Deepdive404

Oh! I got it! Thank you so much for your prompt and clear explanation! The repo duplicator link you shared perfectly addressed my confusion. I really appreciate you taking the time to clarify this - now I understand how repository duplication works on HuggingFace. This was very helpful and I learned something new today!

Thanks again for your guidance!

Deepdive404 changed discussion status to closed Feb 6, 2025

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment