Instructions to use deepseek-ai/DeepSeek-V3-0324 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use deepseek-ai/DeepSeek-V3-0324 with Transformers:

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

pipe = pipeline("text-generation", model="deepseek-ai/DeepSeek-V3-0324", 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("deepseek-ai/DeepSeek-V3-0324", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-V3-0324", trust_remote_code=True, device_map="auto")
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]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use deepseek-ai/DeepSeek-V3-0324 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "deepseek-ai/DeepSeek-V3-0324"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/DeepSeek-V3-0324",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/deepseek-ai/DeepSeek-V3-0324

SGLang

How to use deepseek-ai/DeepSeek-V3-0324 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 "deepseek-ai/DeepSeek-V3-0324" \
    --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": "deepseek-ai/DeepSeek-V3-0324",
		"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 "deepseek-ai/DeepSeek-V3-0324" \
        --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": "deepseek-ai/DeepSeek-V3-0324",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use deepseek-ai/DeepSeek-V3-0324 with Docker Model Runner:
```
docker model run hf.co/deepseek-ai/DeepSeek-V3-0324
```

Regarding the proposal of "keeping it academic", please close the discussion that does not contain academic/technical information.

#41

by likewendy - opened Mar 27, 2025

Discussion

likewendy

Mar 27, 2025

Regarding the proposal of "keeping it academic", please close the discussion that does not contain academic/technical information.
We can post these words in one discussion and collect them together.
Too much praise will lose the value of the hugging face discussion, please stay objective.

Amyww

Mar 28, 2025

gardner

Mar 28, 2025

I appreciate your message @likewendy . Unfortunately, trying to control conversations on the internet is not very realistic in a public forum like the Community posts on Hugging Face. There exists a tradeoff between being open and inclusive; and keeping conversations in a controlled direction.

There are many places for protected higher conversation. These "community posts" are generally open to everyone and contain a wide degree of opinions and "noise". I advise that trying to control the conversation here is a losing proposition that would require effort not commensurate to the benefit.

likewendy

Mar 29, 2025

I appreciate your message @likewendy . Unfortunately, trying to control conversations on the internet is not very realistic in a public forum like the Community posts on Hugging Face. There exists a tradeoff between being open and inclusive; and keeping conversations in a controlled direction.

There are many places for protected higher conversation. These "community posts" are generally open to everyone and contain a wide degree of opinions and "noise". I advise that trying to control the conversation here is a losing proposition that would require effort not commensurate to the benefit.

You are right.
But let me correct you. This is not "control", but suggestion.
My English is so poor that my tone is not tactful enough.
However, people are self-aware, especially those who can log in to hugging face, most of them are well educated.
Once they realize that they may cause trouble, they will do what they should do.

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