MLX Community · First MLX build 24 Dec 2025
Qwen3-Omni-30B-A3B-Instruct
MLX
TextVisionAudio apache-2.0
A 35B parameter any to any model, with 5 MLX builds, from 4-bit to bf16.
- Parameters
- 35B From Hugging Face
- Quantisation
- 4-bit Qwen3-Omni-30B-A3B-Instruct-4bit
- Memory needed
- 27 GB Estimate, unified memory
- Download
- 21.8 GB Weights, estimated
- Downloads
- 1.4K This build, last 30 days
- Task
- Any to any Text, Vision, Audio
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 27 GB at 4-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/Qwen3-Omni-30B-A3B-Instruct-4bitMLX builds
5 on mlx-community| Qwen3-Omni-30B-A3B-Instruct-4bit | 4-bit | 21.8 GB | 27 GB | 1.4K |
|---|---|---|---|---|
| Qwen3-Omni-30B-A3B-Instruct-5bit | 5-bit | 26.1 GB | 32 GB | 107 |
| Qwen3-Omni-30B-A3B-Instruct-6bit | 6-bit | 30.3 GB | 36 GB | 137 |
| Qwen3-Omni-30B-A3B-Instruct-8bit | 8-bit | 38.8 GB | 46 GB | 264 |
| Qwen3-Omni-30B-A3B-Instruct-bf16 | bf16 | 70.5 GB | 81 GB | 188 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.