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

Libraries

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

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

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

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx"

Configure the model in Pi

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

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

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

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx"

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-mlx

Run Hermes

hermes

Atomic Chat

OpenClaw

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

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx"

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-mlx" \
  --custom-provider-id mlx-lm \
  --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-mlx

Qwen3.6-35B-A3B pruned to 19B total / 3.3B active parameters, stored at 2-bit — 6.8× less expert memory than the 8-bit reference. Stock mlx_lm, no custom code.

Niwaki (庭木) are Japan's garden trees, sculpted by meticulous pruning so that every branch serves the form of the whole. This model applies that spirit to a Mixture-of-Experts: every routed expert is individually width-pruned using statistics of its routed tokens over a mixed web/code/chat/reasoning corpus, reconstructed to compensate, then distilled from the full model on the same mixed corpus, and stored at low precision. The result is an ordinary MoE — smaller, same architecture, standard fused kernels.

A paper with the full method and measurements is coming soon.

Benchmarks

Full evaluation protocol: WikiText-2 perplexity over 256 × 2048-token windows; task average over the complete arc_easy, hellaswag, piqa, winogrande, and boolq test sets, paired against the reference.

model	expert storage	wt2 ppl ↓	task avg ↑	retention	generation d2 (avg/min)
reference (8-bit)	34.2 GB	6.76	0.796	100%	0.89 / 0.77
this model	5.0 GB (0.147×)	12.38	0.742	93.3%	0.89 / 0.75

Generation quality is measured, not assumed: an 8-prompt battery (code, reasoning, chat, creative; 600-token sampled generations) scored by bigram diversity.

Model dimensions


total parameters	19B (from 35B)
active parameters per token	~3.3B (from ~3.8B: 2.8B backbone + 8 pruned experts)
layers / routed experts / top-k	40 / 256 / 8
expert intermediate size	256 (from 512)
expert precision	2-bit (group 64)
backbone precision	8-bit (untouched)
whole-model size on disk	~8.0 GB
runs comfortably in	12–16 GB unified memory

Usage (MLX, Apple Silicon)

from mlx_lm import load, generate

model, tokenizer = load("neopolita/Qwen3.6-19B-A3B-Niwaki-2bit-mlx")
print(generate(model, tokenizer, prompt="...", max_tokens=256))

GGUF builds for llama.cpp: Qwen3.6-19B-A3B-Niwaki-2bit-GGUF (UD-Q3K recommended + Q4_K_M).

Notes and limitations

Compression trades quality: this model sits at ~93% of the reference task average and 1.83× its perplexity. Choose the family member that fits your memory budget.
Evaluated text-only on English-web-heavy data; the base model's biases are inherited and rare-domain behaviour of pruned experts is less tested.
Part of the Niwaki family: 27B-A3B-Niwaki-2bit · 11B-A3B-Niwaki-4bit.

Base model by the Qwen team (Apache 2.0); 8-bit MLX conversion by mlx-community; pruning and distillation by the Niwaki project, 2026-08. Method details: paper coming soon.