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

Libraries

How to use microsoft/Florence-2-large with Transformers:

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

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

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("microsoft/Florence-2-large", trust_remote_code=True)
model = AutoModelForMultimodalLM.from_pretrained("microsoft/Florence-2-large", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use microsoft/Florence-2-large with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/Florence-2-large"
# 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-large",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/microsoft/Florence-2-large

SGLang

How to use microsoft/Florence-2-large 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-large" \
    --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-large",
		"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-large" \
        --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-large",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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

masterofaudio2077 commited on 4 days ago

Commit

67061fd

verified ·

1 Parent(s): 21a599d

Fix two transformers-v5 AttributeErrors in Florence-2 remote code

Browse files

transformers v5 changed two things this repo's remote code still assumes the old behavior for (see MIGRATION_GUIDE_V5.md):

1. `forced_bos_token_id` is no longer auto-set as a config instance attribute ('Modeling > Generate'). `Florence2LanguageConfig` read `self.forced_bos_token_id` assuming the old behavior.
2. `additional_special_tokens` was renamed to `extra_special_tokens`. `Florence2Processor.__init__` read the old attribute name, raising AttributeError.

Both fixes are minimal and backward compatible with transformers v4. Related: kijai/ComfyUI-Florence2#187, #199, Comfy-Org/ComfyUI#12277, Lightricks/ComfyUI-LTXVideo#394 all hit the symptom of bug 1. #119 also contains fixes for both but bundles them with a full-repo rewrite that corrupts config.json (swaps in Florence-2-base's architecture dimensions), so it's not safe to merge as-is -- this PR is scoped to just these two AttributeErrors. Supersedes #121, which fixed bug 1 alone and later had bug 2 added as a second commit; this PR bundles both from the start instead.

Files changed (2) hide show

configuration_florence2.py +1 -0
processing_florence2.py +3 -1

configuration_florence2.py CHANGED Viewed

@@ -261,6 +261,7 @@ class Florence2LanguageConfig(PretrainedConfig):
             **kwargs,
         )
         # ensure backward compatibility for BART CNN models
         if self.forced_bos_token_id is None and kwargs.get("force_bos_token_to_be_generated", False):
             self.forced_bos_token_id = self.bos_token_id

             **kwargs,
         )
+        self.forced_bos_token_id = kwargs.get("forced_bos_token_id", None)
         # ensure backward compatibility for BART CNN models
         if self.forced_bos_token_id is None and kwargs.get("force_bos_token_to_be_generated", False):
             self.forced_bos_token_id = self.bos_token_id

processing_florence2.py CHANGED Viewed

@@ -86,7 +86,9 @@ class Florence2Processor(ProcessorMixin):
         tokens_to_add = {
                 'additional_special_tokens': \
-                    tokenizer.additional_special_tokens + \
                     ['<od>', '</od>', '<ocr>', '</ocr>'] + \
                     [f'<loc_{x}>' for x in range(1000)] + \
                     ['<cap>', '</cap>', '<ncap>', '</ncap>','<dcap>', '</dcap>', '<grounding>', '</grounding>', '<seg>', '</seg>', '<sep>', '<region_cap>', '</region_cap>', '<region_to_desciption>', '</region_to_desciption>', '<proposal>', '</proposal>', '<poly>', '</poly>', '<and>']

         tokens_to_add = {
                 'additional_special_tokens': \
+                    (getattr(tokenizer, "extra_special_tokens", None)
+                     if getattr(tokenizer, "extra_special_tokens", None) is not None
+                     else getattr(tokenizer, "additional_special_tokens", [])) + \
                     ['<od>', '</od>', '<ocr>', '</ocr>'] + \
                     [f'<loc_{x}>' for x in range(1000)] + \
                     ['<cap>', '</cap>', '<ncap>', '</ncap>','<dcap>', '</dcap>', '<grounding>', '</grounding>', '<seg>', '</seg>', '<sep>', '<region_cap>', '</region_cap>', '<region_to_desciption>', '</region_to_desciption>', '<proposal>', '</proposal>', '<poly>', '</poly>', '<and>']