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
```

Review of DeepSeek R1 Zero

#24

by nikitayev - opened Apr 21, 2025

Discussion

nikitayev

Apr 21, 2025

You can use it on OpenRouter: https://openrouter.ai/deepseek/deepseek-r1-zero:free

The test task results are available here: https://disk.yandex.ru/d/eNQF9Fe0RtEwxg
The full folder of examples is here: https://disk.yandex.ru/d/iP_f37VTFKm_rA

I experimented with various settings, but the most interesting results came from:

Temperature: 0.6
Top P: 0.95
Top K: 100
Min P: 0.00

Examples with these settings are in the folders starting from:
2 without prompt, lenient conditions\variant 14
to
2 without prompt, lenient conditions\variant 18

The results for variants 14, 15, 16, and 17 are identical.

In fact, these outcomes are on par with those from today’s most advanced AI models.
The results could improve if Top_K could be increased.

Practical experience shows that higher Top_K values lead to better outputs—someone should shout this at all AI interface developers!
For example, in LM Studio, setting Top_K to 500 achieves superior results.

P.S.

My Telegram: https://t.me/Nikitayev
Telegram group for high-tech news: https://t.me/nikitaevai

Original: https://nikitayev.livejournal.com/157013.html

Jzusbsoksibs

Sep 2, 2025

•

edited Sep 2, 2025

Créer un système avancé utilisant plus de 100 algorithmes pour prédire les matchs de football de tous les championnats. Il fournit des prédictions sur les résultats possibles du match. Ces prédictions sont étudiées et confirmées. Il fournit également les meilleures réponses. 2/9/2025 Gentofte-Vangede IF vs Bronshoj BK Fournit des prédictions solides telles que le vainq

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