Instructions to use zhaode/FastVLM-7B-Stage2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zhaode/FastVLM-7B-Stage2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="zhaode/FastVLM-7B-Stage2", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("zhaode/FastVLM-7B-Stage2", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use zhaode/FastVLM-7B-Stage2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zhaode/FastVLM-7B-Stage2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zhaode/FastVLM-7B-Stage2",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/zhaode/FastVLM-7B-Stage2

SGLang

How to use zhaode/FastVLM-7B-Stage2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "zhaode/FastVLM-7B-Stage2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zhaode/FastVLM-7B-Stage2",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "zhaode/FastVLM-7B-Stage2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zhaode/FastVLM-7B-Stage2",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use zhaode/FastVLM-7B-Stage2 with Docker Model Runner:
```
docker model run hf.co/zhaode/FastVLM-7B-Stage2
```

FastVLM-7B-Stage2

Introduction

This is FastVLM-7B-Stage2, 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.

Model's weight: llava-fastvithd_7b_stage2.zip.

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = 'FastVLM-7B-Stage2'
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

git clone https://github.com/alibaba/MNN
cd MNN/transformers/llm/export
python llmexport.py --path /path/to/FastVLM-7B-Stage2 --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}

Downloads last month: 9

Safetensors

Model size

8B params

Tensor type

BF16

Collection including zhaode/FastVLM-7B-Stage2

FastVLM

Collection

FastVLM support transformers load. • 6 items • Updated May 21, 2025