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
```

request for ollama

#17

by ainz - opened Mar 24, 2025

Discussion

ainz

Mar 24, 2025

can you pls add it to ollama ty

smartkit

Mar 25, 2025

ollama run deepseek-ai/DeepSeek-V3-0324
pulling manifest
Error: pull model manifest: file does not exist

wqerrewetw

Mar 25, 2025

•

edited Mar 25, 2025

https://huggingface.co/models?other=base_model:quantized:deepseek-ai/DeepSeek-V3-0324

ollama run hf.co/unsloth/DeepSeek-V3-0324-GGUF:Q2_K
ollama run hf.co/unsloth/DeepSeek-V3-0324-GGUF:Q2_K_XL
ollama run hf.co/unsloth/DeepSeek-V3-0324-GGUF:Q3_K_M
ollama run hf.co/unsloth/DeepSeek-V3-0324-GGUF:Q4_K_M
ollama run hf.co/unsloth/DeepSeek-V3-0324-GGUF:Q5_K_M
ollama run hf.co/unsloth/DeepSeek-V3-0324-GGUF:Q6_K
ollama run hf.co/unsloth/DeepSeek-V3-0324-GGUF:Q8_0


ollama run hf.co/MaziyarPanahi/DeepSeek-V3-0324-GGUF:Q2_K

ollama run hf.co/lmstudio-community/DeepSeek-V3-0324-GGUF:Q3_K_L
ollama run hf.co/lmstudio-community/DeepSeek-V3-0324-GGUF:Q4_K_M
ollama run hf.co/lmstudio-community/DeepSeek-V3-0324-GGUF:Q6_K
ollama run hf.co/lmstudio-community/DeepSeek-V3-0324-GGUF:Q8_0

xinrui-wang1

Mar 25, 2025

ollama run hf.co/lmstudio-community/DeepSeek-V3-0324-GGUF:Q4_K_M

pulling manifest
Error: pull model manifest: 400: The specified repository contains sharded GGUF. Ollama does not support this yet. Follow this issue for more info: https://github.com/ollama/ollama/issues/5245

createthis

Mar 27, 2025

•

edited Mar 30, 2025

https://ollama.com/lordoliver/DeepSeek-V3-0324:671b-q8_0

ollama run lordoliver/DeepSeek-V3-0324:671b-q8_0

EDIT: This runs on my machine, but it keeps talking about Youtube scripts unprompted. Might be a personalized fine tune? An official version would be nice.
EDIT2: It works if you take the Modelfile from the original deepseek-v3 and use that. I think there was a custom prompt lordoliver is using. Here's how to do that (only works if you also have deepseek-v3:671b-q8_0 downloaded):

ollama stop lordoliver/DeepSeek-V3-0324:671b-q8_0
mkdir /data/DeepSeek-V3-0324
cd /data/DeepSeek-V3-0324
ollama show deepseek-v3:671b-q8_0 --modelfile > Modelfile.deepseek
ollama show lordoliver/DeepSeek-V3-0324:671b-q8_0 --modelfile > Modelfile.old
cp Modelfile.deepseek Modelfile
vim Modelfile.old
# copy the FROM line then open Modelfile in a new vim tab
tabe Modelfile
# paste the FROM line from Modelfile.old into Modelfile
ollama create DeepSeek-V3-0324:671b-q8_0 -f ./Modelfile
ollama run DeepSeek-V3-0324:671b-q8_0

EDIT3: It was just pointed out that 0324 is 685B and lordoliver's model is 671b, so is this really 0324? I asked the unsloth community here: https://huggingface.co/unsloth/DeepSeek-V3-0324-GGUF/discussions/8

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