amd · First MLX build 3 Oct 2024
AMD-Llama-135m
MLX
apache-2.0
A 134M parameter model from amd, with 6 MLX builds, from 4-bit to 8-bit.
- Parameters
- 134M From Hugging Face
- Quantisation
- 4-bit AMD-Llama-135m-4bit
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 75 MB Weights, estimated
- Downloads
- 151 This build, last 30 days
- Task
- —
- MLX builds
- 6 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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/AMD-Llama-135m-4bitMLX builds
6 on mlx-community| AMD-Llama-135m-4bit | 4-bit | 75 MB | 3 GB | 151 |
|---|---|---|---|---|
| AMD-Llama-135m-8bit | 8-bit | 143 MB | 3 GB | 14 |
| AMD-Llama-135m-bfloat16 | Unstated | 268 MB | 3 GB | 12 |
| AMD-Llama-135m | Unstated | 268 MB | 3 GB | 11 |
| AMD-Llama-135m-float16 | Unstated | 268 MB | 3 GB | 8 |
| AMD-Llama-135m-float32 | Unstated | 536 MB | 3 GB | 10 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.