Instructions to use QuantFactory/Codepy-Deepthink-3B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantFactory/Codepy-Deepthink-3B-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="QuantFactory/Codepy-Deepthink-3B-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("QuantFactory/Codepy-Deepthink-3B-GGUF", dtype="auto")

llama-cpp-python

How to use QuantFactory/Codepy-Deepthink-3B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="QuantFactory/Codepy-Deepthink-3B-GGUF",
	filename="Codepy-Deepthink-3B.Q2_K.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use QuantFactory/Codepy-Deepthink-3B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf QuantFactory/Codepy-Deepthink-3B-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 QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf QuantFactory/Codepy-Deepthink-3B-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 QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use QuantFactory/Codepy-Deepthink-3B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuantFactory/Codepy-Deepthink-3B-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": "QuantFactory/Codepy-Deepthink-3B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

SGLang

How to use QuantFactory/Codepy-Deepthink-3B-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 "QuantFactory/Codepy-Deepthink-3B-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": "QuantFactory/Codepy-Deepthink-3B-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 "QuantFactory/Codepy-Deepthink-3B-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": "QuantFactory/Codepy-Deepthink-3B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use QuantFactory/Codepy-Deepthink-3B-GGUF with Ollama:
```
ollama run hf.co/QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M
```

Unsloth Studio new

How to use QuantFactory/Codepy-Deepthink-3B-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 QuantFactory/Codepy-Deepthink-3B-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 QuantFactory/Codepy-Deepthink-3B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for QuantFactory/Codepy-Deepthink-3B-GGUF to start chatting

Pi new

How to use QuantFactory/Codepy-Deepthink-3B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use QuantFactory/Codepy-Deepthink-3B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use QuantFactory/Codepy-Deepthink-3B-GGUF with Docker Model Runner:
```
docker model run hf.co/QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M
```

Lemonade

How to use QuantFactory/Codepy-Deepthink-3B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull QuantFactory/Codepy-Deepthink-3B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Codepy-Deepthink-3B-GGUF-Q4_K_M

List all available models

lemonade list

License Incompatibility

by xixi126 - opened Jun 23, 2025

Discussion

xixi126

Jun 23, 2025

•

edited Jun 23, 2025

Hi, I’d like to report a potential license conflict in QuantFactory/Codepy-Deepthink-3B-GGUF. From what I can tell, this model appears to be a quantized version of
meta-llama/Llama-3.2-3B-Instruct, which is licensed under the LLaMA 3.2 Community License . However, the quantized model is currently published under the creativeml-openrail-m License, which might not be compatible due to legal and distribution restrictions from the LLaMA 3.2 license.

⚠️ Key Violations of the LLaMA 3.2 Community License:

Clause 1.b.i – Redistribution and Use:
  • Derivatives must retain the original license
  • Must include a “Built with LLaMA” statement

Clause 1.b.iii – Required Attribution:
  • A “NOTICE” file must be included with:
    "Llama 3.2 is licensed under the LLaMA 3.2 Community License, Copyright © Meta Platforms, Inc."

Clause 1.b.iv – Acceptable Use Policy:
  • No mention or pass-through of Meta’s Acceptable Use Policy, which is mandatory for downstream use

Clause 2 – Additional Commercial Terms:
  • No indication of whether the 700M MAU commercial threshold applies — which makes compliance ambiguous

The CreativeML OpenRAIL-M license:

  • Does not require recipients to accept the LLaMA 3.2 Community License;

  • No strict naming, attribution, or usage requirements

  • No need to pass down an Acceptable Use Policy

By contrast, LLaMA 3.2 imposes non-transferable, license-locked rights. So publishing a LLaMA 3.2-derived model under CreativeML OpenRAIL-M license removes essential obligations that Meta explicitly requires.

🔹 Suggestions for Resolving

To align with the LLaMA 3.2 license:
1. Add this to a NOTICE file or model card:
  > "Llama 3.2 is licensed under the LLaMA 3.2 Community License, Copyright © Meta Platforms, Inc."
2. Add a note that usage must comply with Meta’s Acceptable Use Policy
3. Optionally clarify whether the 700M MAU clause applies or not (if commercial use is intended)
4. Replace the creativeml-openrail-m license reference with the full LLaMA 3.2 Community License text or link
5. Replace the tag of creativeml-openrail-m license reference with LLaMA 3.2 Community License

Hope this helps clarify the situation! Let me know if you have any questions or need help updating the license terms — happy to assist 😊 Let me know if I misunderstood anything — happy to help clarify further!

Thanks for your attention!

xixi126

Jul 10, 2025

Your reply would be much appreciated!

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