swiss-ai · First MLX build 3 Sept 2025
Apertus-8B-Instruct-2509
MLX
Text apache-2.0
A 8.1B parameter text generation model from swiss-ai, with 4 MLX builds, from 4-bit to bf16.
- Parameters
- 8.1B From Hugging Face
- Quantisation
- bf16 Apertus-8B-Instruct-2509-bf16
- Memory needed
- 21 GB Estimate, unified memory
- Download
- 16.1 GB Weights, estimated
- Downloads
- 329 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 21 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/Apertus-8B-Instruct-2509-bf16 --prompt "Hello"MLX builds
4 on mlx-community| Apertus-8B-Instruct-2509-4bit | 4-bit | 4.5 GB | 8 GB | 152 |
|---|---|---|---|---|
| Apertus-8B-Instruct-2509-6bit | 6-bit | 6.5 GB | 10 GB | 53 |
| Apertus-8B-Instruct-2509-8bit | 8-bit | 8.6 GB | 12 GB | 71 |
| Apertus-8B-Instruct-2509-bf16 | bf16 | 16.1 GB | 21 GB | 329 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.