Alibaba Qwen · First MLX build 27 Jun 2026
Qwen-AgentWorld-35B-A3B
MLX
Text apache-2.0
A 35B parameter text generation model from Alibaba Qwen, with 3 MLX builds, from 3-bit to bf16.
- Parameters
- 35B From Hugging Face
- Quantisation
- 4-bit Qwen-AgentWorld-35B-A3B-oQ4
- Memory needed
- 24 GB Estimate, unified memory
- Download
- 19.5 GB Weights, estimated
- Downloads
- 325 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 24 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 · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen-AgentWorld-35B-A3B-oQ4 --prompt "Hello"MLX builds
3 on mlx-community| Qwen-AgentWorld-35B-A3B-oQ3.5 | 3-bit | 15.2 GB | 19 GB | 55 |
|---|---|---|---|---|
| Qwen-AgentWorld-35B-A3B-oQ4 | 4-bit | 19.5 GB | 24 GB | 325 |
| Qwen-AgentWorld-35B-A3B-bf16 | bf16 | 69.3 GB | 80 GB | 74 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.