Instructions to use microsoft/Florence-2-base-ft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/Florence-2-base-ft with Transformers:

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

pipe = pipeline("image-text-to-text", model="microsoft/Florence-2-base-ft", trust_remote_code=True)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("microsoft/Florence-2-base-ft", trust_remote_code=True)
model = AutoModelForImageTextToText.from_pretrained("microsoft/Florence-2-base-ft", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use microsoft/Florence-2-base-ft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/Florence-2-base-ft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/Florence-2-base-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/microsoft/Florence-2-base-ft

SGLang

How to use microsoft/Florence-2-base-ft 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 "microsoft/Florence-2-base-ft" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/Florence-2-base-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "microsoft/Florence-2-base-ft" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/Florence-2-base-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use microsoft/Florence-2-base-ft with Docker Model Runner:
```
docker model run hf.co/microsoft/Florence-2-base-ft
```

Trust remote code requirement - why?

by k8si - opened Jun 20, 2024

Discussion

k8si

Jun 20, 2024

I disabled trust_remote_code in the from_pretrained calls because it is not clear what code this would execute and it seems like a terrible idea to trust arbitrary remote code. I get an error The repository for microsoft/Florence-2-base-ft contains custom code which must be executed to correctly load the model. You can inspect the repository content at https://hf.co/microsoft/Florence-2-base-ft. But there is no additional information about this code provided in the model card.

Would your team be willing to provide a description of the code that must be executed to load this model, as well as a pointer to the python source code that would be downloaded/executed?

CHNtentes

Jun 21, 2024

Anything that is not included in transformers library will require trust_remote_code to execute their modeling_XXX.py. You need python code to know how to init/forward a model.

haipingwu

Jun 21, 2024

hi @k8si , since florence-2 is not yet in the official transformers library, trust_remote_code is needed for loading the model.

clayshoaf

Apr 29, 2025

How can I get around this?

It seems like the code that's running would be in these files:

configuration_florence2.py
modeling_florence2.py
processing_florence2.py

Is that correct? If so, how can I run this with an offline machine?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment