Doctor-Shotgun · First MLX build 24 May 2024
TinyLlama-1.1B-32k
MLX
Text
A 1.1B parameter text generation model from Doctor-Shotgun, with 2 MLX builds, from 4-bit to 8-bit.
- Parameters
- 1.1B From Hugging Face
- Quantisation
- 4-bit JOSIE-TinyLlama-1.1B-32k-base-4bit
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 619 MB Weights, estimated
- Downloads
- 49 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 2 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/JOSIE-TinyLlama-1.1B-32k-base-4bit --prompt "Hello"MLX builds
2 on mlx-community| JOSIE-TinyLlama-1.1B-32k-base-4bit | 4-bit | 619 MB | 3 GB | 49 |
|---|---|---|---|---|
| JOSIE-TinyLlama-1.1B-32k-base-8bit | 8-bit | 1.2 GB | 4 GB | 34 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.