Conversion published by the Outlier team. Not shipped in the Outlier app. Numbers below are from internal experiments, not our release tests. Shipped models: Nano 4B · Lite 9B · Quick 26B · Core 27B.

Run this on your Mac with Outlier — a free macOS app for local MLX inference. Outlier for Mac: free Nano and Lite; Pro is $49 once, on up to 3 Macs · brew install --cask outlier-host/tap/outlier

Qwen3-8B (MLX 4-bit)

MLX 4-bit conversion of Qwen/Qwen3-8B. 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/Qwen3-8B-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.

➡ outlier.host

Other Outlier conversions

License

Inherits from upstream (apache-2.0). See base model card.

Downloads last month: 80

Safetensors

Model size

8B params

Tensor type

U32

BF16

MLX

Hardware compatibility

4-bit

Model tree for Outlier-Ai/Qwen3-8B-MLX-4bit

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Quantized

(514)

this model

Collection including Outlier-Ai/Qwen3-8B-MLX-4bit

Outlier for Apple Silicon (MLX)

Collection

MLX 4-bit builds for M1/M2/M3/M4 Macs. Curated tiers + 19 upstream conversions (DeepSeek, Qwen3, Coder, Gemma, Phi-4, QwQ). • 22 items • Updated Apr 20