Instructions to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF",
	filename="Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn.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 localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS
# Run inference directly in the terminal:
llama-cli -hf localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS
# Run inference directly in the terminal:
llama-cli -hf localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_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 localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS
# Run inference directly in the terminal:
./llama-cli -hf localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_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 localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS
# Run inference directly in the terminal:
./build/bin/llama-cli -hf localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS

Use Docker

docker model run hf.co/localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS

LM Studio
Jan

vLLM

How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-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": "localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS

Ollama
How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with Ollama:
```
ollama run hf.co/localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS
```

Unsloth Studio

How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-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 localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-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 localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF to start chatting

How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_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": "localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-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 localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_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 localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS

Run Hermes

hermes

Docker Model Runner
How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with Docker Model Runner:
```
docker model run hf.co/localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS
```

Lemonade

How to use localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull localweights/Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF:IQ4_XS

Run and chat with the model

lemonade run user.Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn-GGUF-IQ4_XS

List all available models

lemonade list

Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn (GGUF)

35B MoE Qwen3.6-A3B trunk (3B active) + embedded NextN-MTP head, quantized for single-GPU inference.

Trunk: IQ4_XS (imatrix-calibrated)
MTP head: Q8_0 (NextN, kv_only_nextn=true)
File size: ~18.3 GB
Min VRAM: ~21 GB at 32K ctx with KV Q8/Q8

The MTP head is embedded inside the GGUF — no separate drafter file. Recent llama.cpp builds activate it via --spec-type draft-mtp.

Quick start (no spec decode)

llama-server -m Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn.gguf \
  -ngl 999 -fa on -c 32768 --parallel 1 \
  -ctk q8_0 -ctv q8_0 --kv-unified \
  --host 0.0.0.0 --port 8080 --jinja

With speculative decoding (recommended)

Requires llama.cpp built from master after ggml-org/llama.cpp#22673:

llama-server -m Qwen3.6-35B-A3B-MTP-IMAT-IQ4_XS-Q8nextn.gguf \
  -ngl 999 -fa on -c 32768 --parallel 1 \
  -ctk q8_0 -ctv q8_0 --kv-unified \
  --spec-type draft-mtp \
  --spec-draft-n-max 2 \
  --spec-draft-p-min 0.0 \
  --host 0.0.0.0 --port 8080 --jinja

For A3B MoE, smaller --spec-draft-n-max wins — accept rate decays faster with depth than dense models. n_max=2 is the sweet spot. For MoE at deeper chains (n>=3), --spec-draft-p-min 0.0 outperforms p_min=0.75 because per-step sampler overhead from top_k+softmax exceeds the accept-rate savings on this arch.

Benchmarks

Single-stream decode on a 24 GB consumer GPU (RTX 3090 Ti), CUDA build, full-GPU offload, FA on, N=512 generated tokens, 5-warm-run mean.

Production config (ctx=32K, KV Q8/Q8)

config	decode tok/s
no spec (greedy)	~120 (varies)
`--spec-type draft-mtp --spec-draft-n-max 2 --spec-draft-p-min 0.0`	~200 (+67%)

Smaller ctx variant (ctx=8192, KV q4_0/q4_0, p_min=0)

n_max	tok/s
1	189.5
2	198.7 (+24%)
3	191.2
4	167.6

n_max=2 wins; deeper chains regress (MoE accept rate falls faster with depth).

Quant details

Trunk weights: IQ4_XS calibrated with an imatrix derived from a mixed-domain calibration set.
MTP head: kept at Q8_0 (NextN). Sensitive to precision.
KV cache: Q8_0/Q8_0 recommended.

llama.cpp requirements

Build supporting Qwen3.6 NextN MTP (LLM_ARCH_QWEN35MOE_MTP). Merged upstream in #22673.
--spec-draft-p-min 0.75 requires recent llama.cpp that honors p_min on the DRAFT_MTP path.