Instructions to use infly/Infinity-Parser2-Flash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use infly/Infinity-Parser2-Flash with Transformers:

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

pipe = pipeline("image-text-to-text", model="infly/Infinity-Parser2-Flash")
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 AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("infly/Infinity-Parser2-Flash")
model = AutoModelForMultimodalLM.from_pretrained("infly/Infinity-Parser2-Flash")
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?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use infly/Infinity-Parser2-Flash with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "infly/Infinity-Parser2-Flash"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "infly/Infinity-Parser2-Flash",
		"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/infly/Infinity-Parser2-Flash

SGLang

How to use infly/Infinity-Parser2-Flash 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 "infly/Infinity-Parser2-Flash" \
    --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": "infly/Infinity-Parser2-Flash",
		"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 "infly/Infinity-Parser2-Flash" \
        --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": "infly/Infinity-Parser2-Flash",
		"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 infly/Infinity-Parser2-Flash with Docker Model Runner:
```
docker model run hf.co/infly/Infinity-Parser2-Flash
```

Performance degradation issue, please fix

by Obahadir - opened 29 days ago

Discussion

Obahadir

29 days ago

I was having really good results even with flash model 1-2 weeks ago. Today, I can not get same performance even with pro model. How is this possible? Could you reverse what you did to this model ? 😀

zuminghuang

inftech.ai org 28 days ago

We haven't updated any models in the past two weeks, so there’s nothing to roll back just yet 😄. That said, could you share your inference code or configs? We’d be happy to help you debug the issue.

WingedShadow

28 days ago

Hey-,

There is no specific config, we were working on some images to see model performance in demo. After getting really good results, I cloned the model and started to work on my local. Now, I wanted to see some results again in demo for the same image (in my local, it takes time), I could not get very good results. So for demo side, I did not change any in config.

I did not think I was doing something wrong, that's why I checked the last commit, and saw there was an update 13 days ago. I did not check individually yet, but I wanted to ask whether this update could have affected the demo behavior or default settings.

zuminghuang

inftech.ai org 27 days ago

Sorry for the inconvenience. The demo and the model files are based on exactly the same model weights and inference code.

Could you share one of your images? We can help you debug the issue.

WingedShadow

26 days ago

Hey,

We sent an email to your gmail address, including an image and its results.

Looking forward to hearing from you.

zuminghuang changed discussion status to closed 25 days ago

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