---
library_name: transformers
license: apache-2.0
language:
- en
base_model: ds4sd/SmolDocling-256M-preview
pipeline_tag: image-text-to-text
tags:
- mlx
- mlxvlm
---

# ahishamm/SmolDocling-256M-preview-mlx-fp16
The Model [ahishamm/SmolDocling-256M-preview-mlx-fp16](https://huggingface.co/ahishamm/SmolDocling-256M-preview-mlx-fp16) was converted to MLX format from [ds4sd/SmolDocling-256M-preview](https://huggingface.co/ds4sd/SmolDocling-256M-preview)
using mlx-vlm version **0.1.17**.

## Use with mlx
```bash
pip install mlx-vlm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("ahishamm/SmolDocling-256M-preview-mlx-fp16")
prompt = "hello"
if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )
response = generate(model, tokenizer, prompt=prompt, verbose=True)
```