76 lines
1.9 KiB
Markdown
76 lines
1.9 KiB
Markdown
---
|
|
base_model: Qwen/Qwen3.5-9B
|
|
library_name: mlx
|
|
pipeline_tag: image-text-to-text
|
|
tags:
|
|
- mlx
|
|
- qwen3.5
|
|
- vision-language-model
|
|
- quantized
|
|
- 4bit
|
|
license: apache-2.0
|
|
---
|
|
|
|
# Qwen3.5-9B-MLX-4bit
|
|
|
|
This is a quantized MLX version of [Qwen/Qwen3.5-9B](https://huggingface.co/Qwen/Qwen3.5-9B) for Apple Silicon.
|
|
|
|
## Model Details
|
|
|
|
- **Original Model:** [Qwen/Qwen3.5-9B](https://huggingface.co/Qwen/Qwen3.5-9B)
|
|
- **Quantization:** 4-bit (~5.059 bits per weight)
|
|
- **Group Size:** 64
|
|
- **Format:** MLX SafeTensors
|
|
- **Framework:** [mlx-vlm](https://github.com/Blaizzy/mlx-vlm)
|
|
|
|
## Conversion Details
|
|
|
|
This model was converted using `mlx-vlm` with 4-bit quantization.
|
|
|
|
**Conversion command:**
|
|
```bash
|
|
python3 -m mlx_vlm convert \
|
|
--hf-path "Qwen/Qwen3.5-9B" \
|
|
--mlx-path "./mlx_models/Qwen3.5-9B-MLX-4bit" \
|
|
-q --q-bits 4 --q-group-size 64
|
|
```
|
|
|
|
## Important Note
|
|
|
|
A better, more optimized conversion may be available from **@Prince** ([@Blaizzy](https://huggingface.co/Blaizzy)) in the MLX VLM community. Check the [mlx-community](https://huggingface.co/mlx-community) organization for updated versions as official Qwen3.5 support is merged into the main `mlx-vlm` branch.
|
|
|
|
## Usage
|
|
|
|
```python
|
|
from mlx_vlm import load, generate
|
|
|
|
model, processor = load("mlx-community/Qwen3.5-9B-MLX-4bit")
|
|
|
|
output = generate(
|
|
model,
|
|
processor,
|
|
prompt="Describe this image in detail",
|
|
image="path/to/image.jpg",
|
|
max_tokens=200
|
|
)
|
|
print(output)
|
|
```
|
|
|
|
Or from the command line:
|
|
```bash
|
|
mlx_vlm generate \
|
|
--model mlx-community/Qwen3.5-9B-MLX-4bit \
|
|
--prompt "Describe this image" \
|
|
--image path/to/image.jpg \
|
|
--max-tokens 200
|
|
```
|
|
|
|
## Performance
|
|
|
|
- **Disk Size:** ~5.6 GB
|
|
- Runs efficiently on Apple Silicon Macs (M1/M2/M3/M4)
|
|
- Lower memory footprint compared to 8-bit quantization
|
|
|
|
## License
|
|
|
|
This model inherits the [Apache 2.0 license](https://huggingface.co/Qwen/Qwen3.5-9B/blob/main/LICENSE) from the original Qwen3.5-9B model.
|