Text Generation
MLX
Safetensors
qwen3
4-bit precision
4bit
apple-silicon
chat
conversational
edge-ai
function-calling
instruct
local-llm
m1
m2
m3
m4
mac
mac-mini
mac-studio
macbook-air
macbook-pro
macos
metal
mlx-community
mlx-lm
no-cloud
offline
on-device
outlier
outlier-app
private
private-ai
quantized
qwen
reasoning
thinking
tool-use
Instructions to use Outlier-Ai/Qwen3-32B-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Outlier-Ai/Qwen3-32B-MLX-4bit 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("Outlier-Ai/Qwen3-32B-MLX-4bit") 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
- MLX LM
How to use Outlier-Ai/Qwen3-32B-MLX-4bit with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "Outlier-Ai/Qwen3-32B-MLX-4bit"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "Outlier-Ai/Qwen3-32B-MLX-4bit" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Outlier-Ai/Qwen3-32B-MLX-4bit", "messages": [ {"role": "user", "content": "Hello"} ] }' - Atomic Chat
metadata
license: apache-2.0
pipeline_tag: text-generation
library_name: mlx
base_model: Qwen/Qwen3-32B
base_model_relation: quantized
quantized_by: Outlier-Ai
tags:
- 4-bit
- 4bit
- apple-silicon
- chat
- conversational
- edge-ai
- function-calling
- instruct
- local-llm
- m1
- m2
- m3
- m4
- mac
- mac-mini
- mac-studio
- macbook-air
- macbook-pro
- macos
- metal
- mlx
- mlx-community
- mlx-lm
- no-cloud
- offline
- on-device
- outlier
- outlier-app
- private
- private-ai
- quantized
- qwen
- qwen3
- reasoning
- safetensors
- text-generation
- thinking
- tool-use
language:
- en
- zh
- fr
- es
- pt
- de
- it
- ru
- ja
- ko
- ar
- vi
- th
- nl
- pl
widget:
- example_title: General knowledge
messages:
- role: user
content: Explain mixture-of-experts models in plain language.
- example_title: Privacy rationale
messages:
- role: user
content: Three practical reasons to run an LLM locally instead of in the cloud?
- example_title: Summarize
messages:
- role: user
content: >-
Summarize the difference between 4-bit and 8-bit quantization in two
short paragraphs.
Run this on your Mac with Outlier — a free macOS app for local MLX inference.
Qwen3-32B (MLX 4-bit)
MLX 4-bit conversion of Qwen/Qwen3-32B. License and base-model fields inherit from the original — see YAML frontmatter above.
Load with mlx-lm
pip install mlx-lm
python -m mlx_lm.generate --model Outlier-Ai/Qwen3-32B-MLX-4bit --prompt "Hello" --max-tokens 256
What is Outlier?
A free macOS app that runs MLX models locally — no cloud, no API keys, no usage caps.
âž¡ outlier.host
Other Outlier conversions
- Qwen3-Coder-30B-A3B-Instruct (MLX 4-bit) — MLX 4-bit conversion (1,598 downloads)
- Qwen3-14B (MLX 4-bit) — MLX 4-bit conversion (466 downloads)
- Qwen3-8B (MLX 4-bit) — MLX 4-bit conversion (107 downloads)
- Outlier-Core-27B (MLX 4-bit) — MLX 4-bit conversion (55 downloads)
- Outlier-Nano-4B (MLX 4-bit) — MLX 4-bit conversion (67 downloads)
License
Inherits from upstream (apache-2.0). See base model card.