Z.ai · First MLX build 28 Jul 2025
GLM-4.5-Air
MLX
Text mit
A 107B parameter text generation model from Z.ai, with 13 MLX builds, from 2-bit DWQ to bf16.
- Parameters
- 107B From Hugging Face
- Quantisation
- 4-bit GLM-4.5-Air-4bit
- Memory needed
- 70 GB Estimate, unified memory
- Download
- 60.1 GB Weights, estimated
- Downloads
- 775 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 13 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 70 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/GLM-4.5-Air-4bit --prompt "Hello"MLX builds
13 on mlx-community| GLM-4.5-Air-2bit-DWQ | 2-bit DWQ | 33.4 GB | 40 GB | 50 |
|---|---|---|---|---|
| GLM-4.5-Air-2bit | 2-bit | 33.4 GB | 40 GB | 60 |
| GLM-4.5-Air-3bit-DWQ-v2 | 3-bit DWQ | 46.8 GB | 55 GB | 73 |
| GLM-4.5-Air-3bit-DWQ | 3-bit DWQ | 46.8 GB | 55 GB | 57 |
| GLM-4.5-Air-3bit | 3-bit | 46.8 GB | 55 GB | 493 |
| GLM-4.5-Air-4bit | 4-bit | 60.1 GB | 70 GB | 775 |
| GLM-4.5-Air-nvfp4 | 4-bit | 60.1 GB | 70 GB | 58 |
| GLM-4.5-Air-mxfp4 | MXFP4 | 60.1 GB | 70 GB | 53 |
| GLM-4.5-Air-5bit | 5-bit | 73.5 GB | 85 GB | 54 |
| GLM-4.5-Air-6bit | 6-bit | 86.8 GB | 100 GB | 67 |
| GLM-4.5-Air-8bit | 8-bit | 114 GB | 129 GB | 507 |
| GLM-4.5-Air-mxfp8 | MXFP8 | 114 GB | 129 GB | 73 |
| GLM-4.5-Air-bf16 | bf16 | 214 GB | 241 GB | 302 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.