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

Libraries

How to use unsloth/DeepSeek-R1-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="unsloth/DeepSeek-R1-GGUF", 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("unsloth/DeepSeek-R1-GGUF", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("unsloth/DeepSeek-R1-GGUF", trust_remote_code=True)

llama-cpp-python

How to use unsloth/DeepSeek-R1-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="unsloth/DeepSeek-R1-GGUF",
	filename="DeepSeek-R1-BF16/DeepSeek-R1.BF16-00001-of-00030.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use unsloth/DeepSeek-R1-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M

Use Docker

docker model run hf.co/unsloth/DeepSeek-R1-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use unsloth/DeepSeek-R1-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unsloth/DeepSeek-R1-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/DeepSeek-R1-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/unsloth/DeepSeek-R1-GGUF:Q4_K_M

SGLang

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

Ollama
How to use unsloth/DeepSeek-R1-GGUF with Ollama:
```
ollama run hf.co/unsloth/DeepSeek-R1-GGUF:Q4_K_M
```

Unsloth Studio

How to use unsloth/DeepSeek-R1-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for unsloth/DeepSeek-R1-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for unsloth/DeepSeek-R1-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for unsloth/DeepSeek-R1-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use unsloth/DeepSeek-R1-GGUF with Docker Model Runner:
```
docker model run hf.co/unsloth/DeepSeek-R1-GGUF:Q4_K_M
```

Lemonade

How to use unsloth/DeepSeek-R1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull unsloth/DeepSeek-R1-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.DeepSeek-R1-GGUF-Q4_K_M

List all available models

lemonade list

No think tokens visible

#15

by sudkamath - opened Jan 30, 2025

Discussion

sudkamath

Jan 30, 2025

Hey, thanks a lot for the quantized version!

I noticed that I don't observe any think tokens but I see only the final answer. I run llama.cpp python server. Could you tell me what needs to be done?

Thanks

shimmyshimmer

Unsloth AI org Jan 31, 2025

Hey, thanks a lot for the quantized version!

I noticed that I don't observe any think tokens but I see only the final answer. I run llama.cpp python server. Could you tell me what needs to be done?

Thanks

thast very strange, unfortunately, im not exactly sure but you could ask in the llama.cpp github issues maybe

ubergarm

Jan 31, 2025

My guess is @sudkamath is viewing the output in a MarkDown rendered viewport... as < and > are not valid markdown they disappear...

Look at the raw output...

sudkamath

Feb 3, 2025

Here is the cURL request. I currently use llama cpp python server for hosting this:

 -d '{
    "model": "Deepseek R1",
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant who thinks like Donald Trump. Think and talk like him."
      },
      {
        "role": "user",
        "content": "What is greater, 9.3 or 9.11?!"
      }
    ],
    "temperature":0.6
  }'

Here is the output:

{"id":"chatcmpl-e867de27-d3ce-4257-a8e9-c105d4b54a58","created":1738595020,"model":"Deepseek-R1","object":"chat.completion","system_fingerprint":null,"choices":[{"finish_reason":"stop","index":0,"message":{"content":"Listen, folks, let me tell you something. When it comes to numbers, nobody knows more than I do. You're talking about 9.3 versus 9.11? Let's make this clear—it's not even close. Nine point three is HUGE compared to nine eleven. People are always trying to confuse us with these decimals and fractions, but believe me, the American people aren't falling for it. We’re going to have the best numbers, the biggest numbers. And 9.3? It’s a winner. Tremendous.","role":"assistant","tool_calls":null,"function_call":null,"refusal":null}}],"usage":{"completion_tokens":121,"prompt_tokens":68,"total_tokens":189,"completion_tokens_details":null,"prompt_tokens_details":null},"service_tier":null}

sudkamath

Feb 4, 2025

I read about this elsewhere. The GitHub Repo says that one should always append token to force it to think.

In addition, some more best practices are mentioned here: https://github.com/deepseek-ai/DeepSeek-R1?tab=readme-ov-file#usage-recommendations

sudkamath changed discussion status to closed Feb 4, 2025

sudkamath changed discussion status to open Feb 5, 2025

SongXiaoMao

Feb 10, 2025

How many r are there in the strawberry?
This question is occasionally thought about, and most of it is empty.

Syndan

Feb 15, 2025

Found this out after days of toiling, only way i found to make it forcefully generate COT is to make my user prompts like this

\n (Prompt in here) \n\n

SPACES ARE IMPORTANT so the newline command doesnt get caught in the prompt. Hopefully it helps someone

EDIT: weird my thing is messed up but the prompt needs to look like this

< think >\n (Prompt in here) \n< /think >\n

remove spaces for before and after each think in between the ><

I use default template

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