arcee-ai · First MLX build 3 Jun 2025
Homunculus
MLX
Text apache-2.0
A 12B parameter text generation model from arcee-ai, with 4 MLX builds, from 4-bit to bf16.
- Parameters
- 12B From Hugging Face
- Quantisation
- bf16 Homunculus-bf16
- Memory needed
- 30 GB Estimate, unified memory
- Download
- 24.9 GB Weights, estimated
- Downloads
- 27 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 30 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 · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Homunculus-bf16 --prompt "Hello"MLX builds
4 on mlx-community| Homunculus-4bit | 4-bit | 7.0 GB | 10 GB | 20 |
|---|---|---|---|---|
| Homunculus-6bit | 6-bit | 10.1 GB | 14 GB | 24 |
| Homunculus-8bit | 8-bit | 13.2 GB | 17 GB | 19 |
| Homunculus-bf16 | bf16 | 24.9 GB | 30 GB | 27 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.