Instructions to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-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 neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-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 neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-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 neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-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 neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M

Use Docker

docker model run hf.co/neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-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": "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M

Ollama
How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with Ollama:
```
ollama run hf.co/neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M
```
Unsloth Desktop

How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with Docker Model Runner:
```
docker model run hf.co/neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M
```

Lemonade

How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.6-19B-A3B-Niwaki-2bit-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-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 neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Qwen3.6-19B-A3B-Niwaki-2bit-GGUF

GGUF builds of Qwen3.6-19B-A3B-Niwaki-2bit-mlx — Qwen3.6-35B-A3B pruned to 19B total / 3.3B active parameters — for llama.cpp and everything built on it.

Niwaki (庭木): every routed expert individually width-pruned to the neurons its own routed tokens actually use, reconstructed to compensate, then briefly distilled from the full model, and stored at low precision. A paper with the full method is coming soon.

Files

file	size	wt2 ppl (llama.cpp, 512-ctx)
Qwen3.6-19B-A3B-Niwaki-2bit-UD-Q3K.gguf (recommended)	9.0 GB	13.15 ±0.10
Qwen3.6-19B-A3B-Niwaki-2bit-Q4_K_M.gguf	11.4 GB	13.26 ±0.10

Reference Qwen3.6-35B-A3B at Q8_0 measures 6.95 under the identical protocol (llama-perplexity, WikiText-2 test, 512-token windows). These llama.cpp numbers are not directly comparable to the MLX repo's 2048-window benchmarks; the relative standings match across both.

Generation battery (measured on the canonical MLX weights; reference scores 0.89 / 0.78): bigram-diversity avg/min = 0.89 / 0.75 across an 8-prompt code/reasoning/chat/creative battery.

The recommended UD-Q3K build is quantized structure-aware (importance matrices calibrated on the same mixed web/code/chat/reasoning corpus as the model itself), mirroring the artifact's native allocation: the always-active backbone (attention, shared experts, embeddings) is kept at high precision (Q6_K) while the pruned routed experts ride a compact carrier (q3_k, imatrix-guided). It matches or beats uniform Q4_K_M quality at ~20% fewer bytes on this model family.

Model dimensions


total / active parameters	19B / ~3.3B
layers / routed experts / top-k	40 / 256 / 8
expert intermediate size	256 (from 512)
context	as base model
conversion note	speculative-decoding (MTP) draft block not included

Usage

llama-cli -m Qwen3.6-19B-A3B-Niwaki-2bit-UD-Q3K.gguf -p "your prompt" -n 256
# or serve:
llama-server -m Qwen3.6-19B-A3B-Niwaki-2bit-UD-Q3K.gguf

Requires a recent llama.cpp with Qwen3.6 (hybrid linear-attention) support. Canonical benchmarks, method outline, and the MLX-native artifact: Qwen3.6-19B-A3B-Niwaki-2bit-mlx. Family: 27B-2bit · 11B-4bit.

Base model by the Qwen team (Apache 2.0); pruning, distillation, and GGUF builds by the Niwaki project, 2026-08.