Instructions to use richardyoung/Qwen3-Coder-480B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use richardyoung/Qwen3-Coder-480B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="richardyoung/Qwen3-Coder-480B-GGUF",
	filename="IQ2_XS/Qwen_Qwen3-Coder-480B-A35B-Instruct-IQ2_XS-00001-of-00004.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 richardyoung/Qwen3-Coder-480B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS
# Run inference directly in the terminal:
llama-cli -hf richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS
# Run inference directly in the terminal:
llama-cli -hf richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

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 richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS
# Run inference directly in the terminal:
./llama-cli -hf richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

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 richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS
# Run inference directly in the terminal:
./build/bin/llama-cli -hf richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

Use Docker

docker model run hf.co/richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

LM Studio
Jan

vLLM

How to use richardyoung/Qwen3-Coder-480B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "richardyoung/Qwen3-Coder-480B-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": "richardyoung/Qwen3-Coder-480B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

Ollama
How to use richardyoung/Qwen3-Coder-480B-GGUF with Ollama:
```
ollama run hf.co/richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS
```

Unsloth Studio

How to use richardyoung/Qwen3-Coder-480B-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 richardyoung/Qwen3-Coder-480B-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 richardyoung/Qwen3-Coder-480B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for richardyoung/Qwen3-Coder-480B-GGUF to start chatting

How to use richardyoung/Qwen3-Coder-480B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

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": "richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use richardyoung/Qwen3-Coder-480B-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 richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

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 richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

Run Hermes

hermes

Docker Model Runner
How to use richardyoung/Qwen3-Coder-480B-GGUF with Docker Model Runner:
```
docker model run hf.co/richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS
```

Lemonade

How to use richardyoung/Qwen3-Coder-480B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull richardyoung/Qwen3-Coder-480B-GGUF:IQ2_XS

Run and chat with the model

lemonade run user.Qwen3-Coder-480B-GGUF-IQ2_XS

List all available models

lemonade list

Qwen3-Coder-480B-A35B-Instruct GGUF

GGUF quantizations of Qwen/Qwen3-Coder-480B-A35B-Instruct for use with llama.cpp and Ollama.

Model Overview

Qwen3-Coder-480B is Alibaba's most powerful agentic coding model featuring:

480B total parameters with 35B active (MoE architecture)
256K native context (extendable to 1M with YaRN)
Claude Sonnet-level performance on complex coding tasks
Apache 2.0 license - fully open source

Available Quantizations

Quantization	Size	Files	RAM Required	Quality	Description
IQ2_XS	133GB	4	~150GB	Good	Extreme 2-bit, for limited RAM
IQ3_M	218GB	6	~240GB	Better	Balanced 3-bit (coming soon)
IQ4_XS	257GB	7	~280GB	Great	Recommended 4-bit (coming soon)

Quick Start with Ollama

# IQ2_XS quantization
ollama run richardyoung/qwen3-coder:iq2_xs "Write a Python REST API with FastAPI"

# With extended context
ollama run richardyoung/qwen3-coder:iq2_xs --num-ctx 65536 "Analyze this codebase..."

Quick Start with llama.cpp

# Download all IQ2_XS shards
huggingface-cli download richardyoung/Qwen3-Coder-480B-GGUF --include "IQ2_XS/*" --local-dir .

# Run with llama.cpp
./llama-cli -m IQ2_XS/Qwen_Qwen3-Coder-480B-A35B-Instruct-IQ2_XS-00001-of-00004.gguf \
  -c 32768 -n 2048 \
  -p "Write a binary search tree implementation in Python"

System Requirements

Quantization	Minimum RAM	Recommended
IQ2_XS	150GB	192GB unified (M2/M3/M4 Ultra)
IQ3_M	240GB	256GB+
IQ4_XS	280GB	320GB+

Model Capabilities

Complex code generation across all programming languages
Multi-file refactoring and architecture design
Debugging and code analysis
Tool use and function calling
Long-context code understanding
Agentic workflows with planning and execution

Chat Template

<|im_start|>system
You are Qwen3-Coder, an expert AI coding assistant.<|im_end|>
<|im_start|>user
{user_message}<|im_end|>
<|im_start|>assistant
{assistant_response}<|im_end|>

Credits

Original Model: Qwen/Qwen3-Coder-480B-A35B-Instruct by Alibaba
GGUF Quantization: bartowski
Distribution: Richard Young (deepneuro.ai)

License

Apache 2.0 - Free for commercial and personal use.

Downloads last month: 6

GGUF

Model size

480B params

Architecture

qwen3moe

Hardware compatibility

2-bit

Model tree for richardyoung/Qwen3-Coder-480B-GGUF

Base model

Qwen/Qwen3-Coder-480B-A35B-Instruct

Quantized

(38)

this model

richardyoung
/

Qwen3-Coder-480B-GGUF