mlxradar
Alibaba Qwen · First MLX build 31 Jul 2025 · In Core AI since 8 Jun 2026

Qwen3-Coder-30B-A3B-Instruct

MLXCore AI Text apache-2.0 Trending

A 31B parameter text generation model from Alibaba Qwen, with 15 MLX builds, from 3-bit to bf16 and Apple Core AI export recipes for macOS.

Show details for
Hugging Face Apple recipe
Parameters
31B
From Hugging Face
Quantisation
4-bit
Qwen3-Coder-30B-A3B-Instruct-4bit
Memory needed
22 GB
Estimate, unified memory
Download
17.2 GB
Weights, estimated
Downloads
19K
This build, last 30 days
Task
Text generation
Text
MLX builds
15
On mlx-community
Devices
Mac
Apple silicon

Memory

Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.

Run it

Terminal · mlx-lm
$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen3-Coder-30B-A3B-Instruct-4bit --prompt "Hello"

MLX builds

15 on mlx-community
Click a column to sort. Smallest size and most downloads are highlighted.
Qwen3-Coder-30B-A3B-Instruct-3bit 3-bit 13.4 GB 17 GB 1.2K
Qwen3-Coder-30B-A3B-Instruct-4bit 4-bit 17.2 GB 22 GB 19K
Qwen3-Coder-30B-A3B-Instruct-4bit-dwq-v2 4-bit DWQ 17.2 GB 22 GB 1.2K
Qwen3-Coder-30B-A3B-Instruct-4bit-DWQ 4-bit DWQ 17.2 GB 22 GB 759
Qwen3-Coder-30B-A3B-Instruct-5bit 5-bit 21.0 GB 26 GB 255
Qwen3-Coder-30B-A3B-Instruct-6bit 6-bit 24.8 GB 30 GB 1.3K
Qwen3-Coder-30B-A3B-Instruct-6bit-DWQ-lr3e-7 6-bit DWQ 24.8 GB 30 GB 622
Qwen3-Coder-30B-A3B-Instruct-6bit-DWQ-lr9e-8 6-bit DWQ 24.8 GB 30 GB 130
Qwen3-Coder-30B-A3B-Instruct-8bit 8-bit 32.4 GB 39 GB 1.4K
Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr5e-8 8-bit DWQ 32.4 GB 39 GB 249
Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr9e8 8-bit DWQ 32.4 GB 39 GB 194
Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr2e7 8-bit DWQ 32.4 GB 39 GB 179
Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr1e-6 8-bit DWQ 32.4 GB 39 GB 146
Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr4e7 8-bit DWQ 32.4 GB 39 GB 145
Qwen3-Coder-30B-A3B-Instruct-bf16 bf16 61.1 GB 71 GB 607

Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.

Similar models

Qwen3-30B-A3B
Alibaba Qwen · 31B · apache-2.0
MLX
Quant3-bit / 4-bit … Memory17 GB Downloads10K Added28 Apr 2025
Qwen3-30B-A3B-Instruct-2507
Alibaba Qwen · 31B · apache-2.0
MLX
Quant4-bit / 5-bit … Memory22 GB Downloads7.3K Added29 Jul 2025
Qwen3-30B-A3B-Thinking-2507
Alibaba Qwen · 31B · apache-2.0
MLX
Quant4-bit / bf16 Memory22 GB Downloads253 Added30 Jul 2025
Qwen3-32B
Alibaba Qwen · 33B · apache-2.0
MLX
Quant3-bit / 4-bit … Memory19 GB Downloads2.4K Added28 Apr 2025

Changelog

  1. Core AI macOS. Apple listed qwen3-coder-30b-a3b-instruct in its registry.
  2. MLX 4-bit DWQ. Qwen3-Coder-30B-A3B-Instruct-4bit-dwq-v2 published on mlx-community.
  3. MLX 6-bit DWQ. Qwen3-Coder-30B-A3B-Instruct-6bit-DWQ-lr9e-8 published on mlx-community.
  4. MLX 6-bit DWQ. Qwen3-Coder-30B-A3B-Instruct-6bit-DWQ-lr3e-7 published on mlx-community.
  5. MLX 4-bit DWQ. Qwen3-Coder-30B-A3B-Instruct-4bit-DWQ published on mlx-community.
  6. MLX 8-bit DWQ. Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr1e-6 published on mlx-community.
  7. MLX 8-bit DWQ. Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr5e-8 published on mlx-community.
  8. MLX 8-bit DWQ. Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr9e8 published on mlx-community.