Instructions to use deepseek-ai/DeepSeek-R1-Zero with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use deepseek-ai/DeepSeek-R1-Zero with Transformers:

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

pipe = pipeline("text-generation", model="deepseek-ai/DeepSeek-R1-Zero", 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-R1-Zero", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-R1-Zero", 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 deepseek-ai/DeepSeek-R1-Zero with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "deepseek-ai/DeepSeek-R1-Zero"
# 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-R1-Zero",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/deepseek-ai/DeepSeek-R1-Zero

SGLang

How to use deepseek-ai/DeepSeek-R1-Zero 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-R1-Zero" \
    --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-R1-Zero",
		"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-R1-Zero" \
        --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-R1-Zero",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use deepseek-ai/DeepSeek-R1-Zero with Docker Model Runner:
```
docker model run hf.co/deepseek-ai/DeepSeek-R1-Zero
```

readability and language mixing issues

#21

by unakar - opened Jan 30, 2025

Discussion

unakar

Jan 30, 2025

Does anyone have evidence to prove that Deepseek R1 Zero faces readability and language mixing issues?
I tested many samples and found its output to be quite normal.

Hassankhwileh

Feb 3, 2025

am facing same issue mixing language

G-Nada

Feb 7, 2025

am facing same issue mixing language

Would you mind sharing your log? I am quite interested in when mixing happens.

henryhe0123

Feb 9, 2025

I'm curious about this too, really appreciate any help

blobba

Feb 13, 2025

@unakar examples?

JohnSamson

Feb 15, 2025

Are you people blind to what's written or are you just too stupid to read? In the introduction, it is clearly stated that DeepSeek-R1-Zero has issues that you illiterates are referring to. To quote: 'However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing.' And if you can manage to read a bit further, which might be challenging for some of you it says that DeepSeek-R1 addresses those issues.

ayymen

Dec 23, 2025

@JohnSamson This is "deepseek-ai/DeepSeek-R1-Zero", the model with the aforementioned issue, not "deepseek-ai/DeepSeek-R1". Learn to read first before insulting random people on the internet.

16dvnk

Dec 24, 2025

@JohnSamson The people are saying that there are no issues, when deepseek says so, but it seems you can't read.

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