Instructions to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound",
	filename="Qwen3-Coder-480B-A35B-Instruct-160x15B-Q2_K_S-00001-of-00004.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S
# Run inference directly in the terminal:
llama-cli -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S
# Run inference directly in the terminal:
llama-cli -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

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 Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S
# Run inference directly in the terminal:
./llama-cli -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

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 Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

Use Docker

docker model run hf.co/Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

LM Studio
Jan
Ollama
How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with Ollama:
```
ollama run hf.co/Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S
```

Unsloth Studio

How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound 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 Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound 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 Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound to start chatting

How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

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": "Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

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 Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

Run Hermes

hermes

Docker Model Runner
How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with Docker Model Runner:
```
docker model run hf.co/Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S
```

Lemonade

How to use Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Intel/Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound:Q2_K_S

Run and chat with the model

lemonade run user.Qwen3-Coder-480B-A35B-Instruct-gguf-q2ks-mixed-AutoRound-Q2_K_S

List all available models

lemonade list

Bits for embedding / lm-head / non expert layers

by sokann - opened Aug 12, 2025

Discussion

sokann

Aug 12, 2025

Embedding layer and lm-head layer are fallback to 8 bits and non expert layers are fallback to 4 bits.

Hi, can you explain a bit about the choices? Asking because I see others typically going with slightly higher bits for the non expert layers, compared to the embedding / lm-head layers.

For example, the IQ2_K quant from https://huggingface.co/ubergarm/Qwen3-Coder-480B-A35B-Instruct-GGUF uses:

# Attention
blk\..*\.attn_q.*=iq6_k
blk\..*\.attn_k.*=q8_0
blk\..*\.attn_v.*=q8_0
blk\..*\.attn_output.*=iq6_k

# Routed Experts
blk\..*\.ffn_down_exps\.weight=iq2_kl
blk\..*\.ffn_(gate|up)_exps\.weight=iq2_k

# Non-Repeating Layers
token_embd\.weight=iq4_k
output\.weight=iq6_k

wenhuach

Intel org Aug 12, 2025

We have not explored the mixed-bit configuration, our choices are primarily based on prior experience. You may be able to use our library to produce an even better model, or find more effective settings in other repositories. Our open-sourced algorithm differs from the official implementation. We are working on an even better version, stay tuned!

For the embedding layer and LM head, given that this model is extremely large and the first/last layer is particularly important, we choose to keep them at higher bit precision. Although we have observed that the embedding layer can be quantized to very low bits, we prefer higher bits to preserve accuracy.

For the expert modules and other related components, since experts account for the majority of the model parameters, they must be quantized to lower bits to save memory. In contrast, shared expert layers or routed experts contain relatively few parameters, so we keep them at higher precision. A similar strategy can be found in some open-source models like openai gpt-oss model that only quantize MoE parameters.

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