4-bit Open on Hugging Face
Parameters
108B
Weights (est.)
≈ 61.9 GB
Mac memory (est.)
96 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-vlm on an Apple silicon Mac:
pip install mlx-vlmpython -m mlx_vlm.generate --model mlx-community/GLM-4.6V-nvfp4 --image photo.jpg --prompt "Describe this image"
Details
- Base model
- zai-org/GLM-4.6V (quantized)
- Vendor
- zai-org
- Task
- Vision + text
- Quantisation
- 4-bit
- Licence
- mit
- Downloads
- 54 in the last 30 days · 560 all time · 0 likes
- Published
- 2026-03-05 · updated 2026-03-05 (8mo ago)
- Languages
- zh, en
- Tags
- glm4v_moe
Other MLX builds of zai-org/GLM-4.6V
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| GLM-4.6V-nvfp4 (this one) | 4-bit | ≈ 61.9 GB | 96 GB+ | 54 |
| GLM-4.6V-mxfp4 | MXFP4 | ≈ 61.9 GB | 96 GB+ | 110 |
| GLM-4.6V-6bit | 6-bit | — | — | 71 |