README.md

---
language: en
license: cdla-permissive-2.0
pipeline_tag: image-text-to-text
library_name: mlx
tags:
  - mlx
  - mlx-vlm
  - idefics3
  - quantized
  - 8-bit
base_model: docling-project/CodeFormulaV2
base_model_relation: quantized
datasets:
  - ds4sd/SynthFormulaNet
  - ds4sd/SynthCodeNet
---

# CodeFormulaV2-mlx-q8

8-bit quantized MLX conversion of [docling-project/CodeFormulaV2](https://huggingface.co/docling-project/CodeFormulaV2), produced with `mlx_vlm.convert --quantize --q-bits 8`. The text decoder linear layers are quantized to 8 bits per weight; the vision encoder stays at bf16 (the mlx-vlm convention via `skip_multimodal_module`).

Architecture, training data, and intended use are described on the upstream model page. This repo only re-encodes the weights at lower precision; no retraining or modification of behaviour was performed.

A bf16 variant is also available: [`mlx-community/CodeFormulaV2-mlx-bf16`](https://huggingface.co/mlx-community/CodeFormulaV2-mlx-bf16).

## Usage

```bash
pip install mlx-vlm
```

```python
from mlx_vlm import load, generate
from mlx_vlm.prompt_utils import apply_chat_template

model, processor = load("mlx-community/CodeFormulaV2-mlx-q8")
prompt = apply_chat_template(processor, model.config, "<formula>", num_images=1)
result = generate(
    model, processor,
    prompt=prompt,
    image="path/to/image.png",
    temperature=0.0,
)
print(result.text)
```

Use `"<formula>"` as the prompt for a math-expression image, `"<code>"` for a code-block image, per the upstream model card.

## License and attribution

This is a derivative of [docling-project/CodeFormulaV2](https://huggingface.co/docling-project/CodeFormulaV2), redistributed under the same [Community Data License Agreement – Permissive 2.0 (CDLA-Permissive-2.0)](https://cdla.dev/permissive-2-0/).

Please cite the upstream work:

```bibtex
@techreport{Docling,
  author = {Deep Search Team},
  month = {8},
  title = {{Docling Technical Report}},
  url = {https://arxiv.org/abs/2408.09869},
  eprint = {2408.09869},
  year = {2024}
}
```