Aratako · First MLX build 11 Dec 2024
sarashina2.1-1b-sft
MLX
Text other
A 1.4B parameter text generation model from Aratako, with 3 MLX builds, from 4-bit to 8-bit.
- Parameters
- 1.4B From Hugging Face
- Quantisation
- 4-bit sarashina2.1-1b-sft-4bit
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 792 MB Weights, estimated
- Downloads
- 25 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 3 GB at 4-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/sarashina2.1-1b-sft-4bit --prompt "Hello"MLX builds
3 on mlx-community| sarashina2.1-1b-sft-4bit | 4-bit | 792 MB | 3 GB | 25 |
|---|---|---|---|---|
| sarashina2.1-1b-sft-8bit | 8-bit | 1.5 GB | 4 GB | 22 |
| sarashina2.1-1b-sft | Unstated | 2.8 GB | 6 GB | 20 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.