Instructions to use SirSahOl/Qwen3-0.6B-chat-mlx-16bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SirSahOl/Qwen3-0.6B-chat-mlx-16bit 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("SirSahOl/Qwen3-0.6B-chat-mlx-16bit")

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 SirSahOl/Qwen3-0.6B-chat-mlx-16bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "SirSahOl/Qwen3-0.6B-chat-mlx-16bit"

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": "SirSahOl/Qwen3-0.6B-chat-mlx-16bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use SirSahOl/Qwen3-0.6B-chat-mlx-16bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "SirSahOl/Qwen3-0.6B-chat-mlx-16bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "SirSahOl/Qwen3-0.6B-chat-mlx-16bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "SirSahOl/Qwen3-0.6B-chat-mlx-16bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use SirSahOl/Qwen3-0.6B-chat-mlx-16bit 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 "SirSahOl/Qwen3-0.6B-chat-mlx-16bit"

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 SirSahOl/Qwen3-0.6B-chat-mlx-16bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use SirSahOl/Qwen3-0.6B-chat-mlx-16bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "SirSahOl/Qwen3-0.6B-chat-mlx-16bit"

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 "SirSahOl/Qwen3-0.6B-chat-mlx-16bit" \
  --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"

Qwen3-0.6B-chat-mlx-16bit / README.md

SirSahOl

Upload 16-bit MLX conversion of Qwen/Qwen3-0.6B

e9fdfd7 verified 7 days ago

preview code

Raw

History Blame Contribute Delete

3.78 kB

	---
	library_name: mlx
	pipeline_tag: text-generation
	license: apache-2.0
	base_model: Qwen/Qwen3-0.6B
	tags:
	- mlx
	- safetensors
	- conversational
	---

	# Qwen3-0.6B-mlx-16bit

	> 16-bit MLX conversion of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B) for Apple Silicon.

	Converted by: [SirSahOl](https://huggingface.co/SirSahOl)
	Source model: [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B)
	Framework: [MLX](https://github.com/ml-explore/mlx) by Apple
	Quantization: 16-bit
	Format: safetensors
	License: apache-2.0

	---

	## Quick Start

	### Installation

	```bash
	pip install mlx-lm
	```

	### CLI Usage

	```bash
	# Chat interactively
	mlx_lm.chat --model SirSahOl/Qwen3-0.6B-chat-mlx-16bit

	# Generate text
	mlx_lm.generate --model SirSahOl/Qwen3-0.6B-chat-mlx-16bit --prompt "Your prompt here"
	```

	### Python Usage

	```python
	from mlx_lm import load, generate

	model, tokenizer = load("SirSahOl/Qwen3-0.6B-chat-mlx-16bit")
	response = generate(model, tokenizer, prompt="Your prompt here", max_tokens=256)
	print(response)
	```

	---

	## Performance Benchmarks

	\| Metric \| 4-bit \| 8-bit \| 16-bit \| \|--------\|--------\|--------\|--------\|\| Tokens/sec \| 114.11 \| 70.97 \| 42.97 \| \| TTFT \| 8.77 ms \| 14.09 ms \| 23.28 ms \| \| Peak Memory \| 449.8 MB \| 393.5 MB \| 171.6 MB \|
	> Benchmarked on Apple M1 with 8GB unified memory.
	> Average over 5 runs with 256 max tokens.

	---

	## Who Should Use This?

	\| Your Hardware \| Recommended Quantization \|
	\|---------------\|--------------------------\|
	\| M1/M2 (8GB) \| 4-bit — Best balance of quality and memory usage \|
	\| M1/M2 Pro/Max (16-32GB) \| 8-bit — Higher quality with reasonable memory \|
	\| M2/M3/M4 Ultra (64GB+) \| 16-bit — Full precision, no quality loss \|

	General guidance:
	- Use 4-bit if you want to run this model alongside other applications
	- Use 8-bit if you have the memory and want better quality
	- Use 16-bit for research, evaluation, or if memory isn't a concern

	---

	## Other Quantization Variants

	\| Variant \| Link \|
	\|---------\|------\|
	\| 4-bit \| [SirSahOl/Qwen3-0.6B-chat-mlx-4bit](https://huggingface.co/SirSahOl/Qwen3-0.6B-chat-mlx-4bit) \|
	\| 8-bit \| [SirSahOl/Qwen3-0.6B-chat-mlx-8bit](https://huggingface.co/SirSahOl/Qwen3-0.6B-chat-mlx-8bit) \|
	\| 16-bit \| [SirSahOl/Qwen3-0.6B-chat-mlx-16bit](https://huggingface.co/SirSahOl/Qwen3-0.6B-chat-mlx-16bit) \|

	---

	## Conversion Details

	\| Property \| Value \|
	\|----------\|-------\|
	\| Source Model \| [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B) \|
	\| Quantization \| 16-bit \|
	\| mlx-lm Version \| 0.31.3 \|
	\| Conversion Time \| 6.7s \|
	\| Output Size \| 1.1 GB \|
	\| Date \| 2026-09-10T17:45:49.821886+00:00 \|

	### Reproduction

	To reproduce this conversion:

	```bash
	pip install mlx-lm==0.31.3
	python3 -m mlx_lm.convert --hf-path Qwen/Qwen3-0.6B --mlx-path output/Qwen3-0.6B-mlx-16bit
	```

	---

	## Limitations & Known Issues

	- Performance may degrade with very long contexts (>8K tokens) at lower quantization levels.
	- This is a weight-only conversion; the model architecture and behavior are inherited from the source model.
	- Quantization introduces a small quality loss compared to the original model. Lower bit counts = more loss.
	- This model requires Apple Silicon (M1 or later) to run with MLX.

	---

	## License

	This model conversion inherits the license of the source model: apache-2.0.

	See the [original model card](https://huggingface.co/Qwen/Qwen3-0.6B) for full license details.

	---

	## Changelog

	\| Version \| Date \| Changes \|
	\|---------\|------\|---------\|
	\| v1.0 \| 2026-09-10 \| Initial conversion \|

	---

	Converted with [MLX Foundry](https://github.com/SirSahOl/mlx-foundry) — a professional pipeline for converting models to Apple MLX format.