---
base_model: Cloudflare/clef
library_name: mlx
pipeline_tag: image-text-to-text
tags:
- mlx
- multimodal
- decision-model
- 8bit
license: apache-2.0
---

# clef-mlx-8bit

This model is a 8-bit quantized version of [Cloudflare/clef](https://huggingface.co/Cloudflare/clef) using the [MLX](https://github.com/ml-explore/mlx) framework.

## Model Details
- **Base Model:** [Cloudflare/clef](https://huggingface.co/Cloudflare/clef)
- **Architecture:**  + Joint Schema Head
- **Quantization:** 8-bit affine (group size: 64)
- **Effective Bits Per Weight:** 8.627 bpw
- **Model Size:** ~28 GB
- **License:** Apache-2.0

## Files Included
- Quantized backbone weights (, )
- Joint Schema Head weights (, )
- Model execution code ()
- Processor & Tokenizer configs (, , , )
- Model Configuration (, )

## Usage with MLX