Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mlx-community/FastVLM-0.5B-bf16
|
| 4 |
+
language:
|
| 5 |
+
- en
|
| 6 |
+
tags:
|
| 7 |
+
- mlx
|
| 8 |
+
- ffai
|
| 9 |
+
- quantized
|
| 10 |
+
- 4bit
|
| 11 |
+
- affine
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# FastVLM-0.5B-4bit
|
| 15 |
+
|
| 16 |
+
4-bit affine quantization of [mlx-community/FastVLM-0.5B-bf16](https://huggingface.co/mlx-community/FastVLM-0.5B-bf16), produced with [FFAI](https://github.com/thewafflehaus/FFAI) 0.1.0's `ffai convert` (mlx-affine format, `group_size=64`).
|
| 17 |
+
|
| 18 |
+
## Conversion
|
| 19 |
+
|
| 20 |
+
```bash
|
| 21 |
+
ffai convert mlx-community/FastVLM-0.5B-bf16 --bits 4 \
|
| 22 |
+
--upload-repo ekryski/FastVLM-0.5B-4bit
|
| 23 |
+
```
|
| 24 |
+
|
| 25 |
+
## See also
|
| 26 |
+
|
| 27 |
+
- [FFAI](https://github.com/thewafflehaus/FFAI) — fast Apple Silicon LLM inference. `Model.load("ekryski/FastVLM-0.5B-4bit")` runs this checkpoint end-to-end.
|
| 28 |
+
- [FFAI quickstart](https://github.com/thewafflehaus/FFAI/blob/main/documentation/quickstart.md)
|
| 29 |
+
- [FFAI quantization docs](https://github.com/thewafflehaus/FFAI/blob/main/documentation/quantization.md)
|