MLX Community · First MLX build 30 Apr 2024
llava-llama-3-8b-v1_1
MLX
TextVision
A 8.4B parameter vision and text model, with 2 MLX builds, from 4-bit to 8-bit.
- Parameters
- 8.4B From Hugging Face
- Quantisation
- 8-bit llava-llama-3-8b-v1_1-8bit
- Memory needed
- 13 GB Estimate, unified memory
- Download
- 9.4 GB Weights, estimated
- Downloads
- 48 This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 2 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 13 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-vlm$ pip install -U mlx-vlm
$ python -m mlx_vlm.generate --model mlx-community/llava-llama-3-8b-v1_1-8bit --image photo.jpg --prompt "Describe this image"MLX builds
2 on mlx-community| llava-llama-3-8b-v1_1-4bit | 4-bit | 5.5 GB | 9 GB | 46 |
|---|---|---|---|---|
| llava-llama-3-8b-v1_1-8bit | 8-bit | 9.4 GB | 13 GB | 48 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.