canopylabs · First MLX build 2 Apr 2025
orpheus-3b-0.1-pretrained
MLX
Text apache-2.0
A 3.3B parameter text generation model from canopylabs, with 4 MLX builds, from 4-bit to bf16.
- Parameters
- 3.3B From Hugging Face
- Quantisation
- 8-bit orpheus-3b-0.1-pretrained-8bit
- Memory needed
- 6 GB Estimate, unified memory
- Download
- 3.5 GB Weights, estimated
- Downloads
- 32 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 6 GB at 8-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/orpheus-3b-0.1-pretrained-8bit --prompt "Hello"MLX builds
4 on mlx-community| orpheus-3b-0.1-pretrained-4bit | 4-bit | 1.9 GB | 5 GB | 29 |
|---|---|---|---|---|
| orpheus-3b-0.1-pretrained-6bit | 6-bit | 2.7 GB | 5 GB | 23 |
| orpheus-3b-0.1-pretrained-8bit | 8-bit | 3.5 GB | 6 GB | 32 |
| orpheus-3b-0.1-pretrained-bf16 | bf16 | 6.6 GB | 10 GB | 19 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.