---
title: NuExtract3-mlx-nvfp4
canonical_url: "https://www.modelscope.ai/models/numind/NuExtract3-mlx-nvfp4"
md_url: "https://www.modelscope.ai/models/numind/NuExtract3-mlx-nvfp4.md"
repository: numind/NuExtract3-mlx-nvfp4
last_updated: 2026-05-22
model_type:
  - qwen3_5
architectures:
  - Qwen3_5ForConditionalGeneration
base_model:
  - numind/NuExtract3
base_model_relation: quantized
parameters: 1.1B
tensor_type:
  - BF16
  - U8
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 168
stars: 0
tags:
  - mlx
  - mlx-vlm
  - quantized
---

# NuExtract3-mlx-nvfp4

> NuExtract3-mlx-nvfp4 - An open-source model by numind on ModelScope. numind/NuExtract3-mlx-nvfp4

numind/NuExtract3-mlx-nvfp4 is a 1.1B-parameter machine learning model on ModelScope. derived from numind/NuExtract3.

- **Repository**: numind/NuExtract3-mlx-nvfp4
- **Parameters**: 1.1B
- **Base model**: numind/NuExtract3
- **Tags**: mlx, mlx-vlm, quantized
- **Downloads**: 168
- **Stars**: 0
- **Last updated**: 2026-05-22

Source: https://www.modelscope.ai/models/numind/NuExtract3-mlx-nvfp4

---

# numind/NuExtract3-mlx-nvfp4

MLX quantized version of `numind/NuExtract3`.

## Quantization

- mode: `nvfp4`
- group_size: `16`
- bits: `4`
- dtype: `source/default`
- quant_predicate: `none`

## Usage

```bash
pip install -U mlx-vlm
mlx_vlm.generate --model numind/NuExtract3-mlx-nvfp4 --prompt "your prompt"
```