HiDream-ai · First MLX build 9 May 2026
HiDream-O1-Image-Dev
MLX
Image mit
A 8.8B parameter text to image model from HiDream-ai, with 3 MLX builds, from 6-bit to bf16.
- Parameters
- 8.8B From Hugging Face
- Quantisation
- bf16 HiDream-O1-Image-Dev-mlx-bf16
- Memory needed
- 22 GB Estimate, unified memory
- Download
- 17.5 GB Weights, estimated
- Downloads
- 284 This build, last 30 days
- Task
- Text to image Image
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 22 GB at bf16
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/HiDream-O1-Image-Dev-mlx-bf16MLX builds
3 on mlx-community| HiDream-O1-Image-Dev-mlx-q6 | 6-bit | 8.0 GB | 11 GB | 76 |
|---|---|---|---|---|
| HiDream-O1-Image-Dev-mlx-q8 | 8-bit | 10.0 GB | 14 GB | 51 |
| HiDream-O1-Image-Dev-mlx-bf16 | bf16 | 17.5 GB | 22 GB | 284 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.