---
title: FastVLM-7B-Stage3
canonical_url: "https://www.modelscope.cn/models/zhaode/FastVLM-7B-Stage3"
md_url: "https://www.modelscope.cn/models/zhaode/FastVLM-7B-Stage3.md"
repository: zhaode/FastVLM-7B-Stage3
last_updated: 2025-05-21
license: other
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
model_type:
  - llava_qwen2
architectures:
  - LlavaQwen2ForCausalLM
parameters: 7.8B
tensor_type:
  - BF16
  - I64
library_name:
  - transformer
  - safetensors
  - other
frameworks:
  - other
language:
  - en
downloads: 39
stars: 0
tags:
  - multimodal
---

# FastVLM-7B-Stage3

> FastVLM-7B-Stage3 - zhaode 在 ModelScope 开源的模型。This is FastVLM-7B-Stage3, a multimodal language model that can understand things visually, being agentic, understand long videos and capture events, and generate structured outputs.

zhaode/FastVLM-7B-Stage3 是 ModelScope 魔搭社区上的 7.8B 参数image-text-to-text模型，采用 other 许可。

- **Repository**: zhaode/FastVLM-7B-Stage3
- **License**: other
- **Tasks**: image-text-to-text
- **Parameters**: 7.8B
- **Tags**: multimodal
- **Downloads**: 39
- **Stars**: 0
- **Last updated**: 2025-05-21

Source: https://www.modelscope.cn/models/zhaode/FastVLM-7B-Stage3

---

# FastVLM-7B-Stage3

## Introduction

This is FastVLM-7B-Stage3, a multimodal language model that can understand things visually, being agentic, understand long videos and capture events, and generate structured outputs.

This model is exported from Github [apple/ml-fastvlm](https://github.com/apple/ml-fastvlm).

Model's weight: [llava-fastvithd_7b_stage3.zip](https://ml-site.cdn-apple.com/datasets/fastvlm/llava-fastvithd_7b_stage3.zip).


### Usage
```python
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = 'FastVLM-7B-Stage3'
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True, use_fast=False)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype='auto', trust_remote_code=True)
```

### Export to MNN
```python
git clone https://github.com/alibaba/MNN
cd MNN/transformers/llm/export
python llmexport.py --path /path/to/FastVLM-7B-Stage3 --export mnn
```


## Citation

If you find our work helpful, feel free to give us a cite.

```
@InProceedings{fastvlm2025,
  author = {Pavan Kumar Anasosalu Vasu, Fartash Faghri, Chun-Liang Li, Cem Koc, Nate True, Albert Antony, Gokul Santhanam, James Gabriel, Peter Grasch, Oncel Tuzel, Hadi Pouransari},
  title = {FastVLM: Efficient Vision Encoding for Vision Language Models},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  month = {June},
  year = {2025},
}{2023}
```