Instructions to use TIGER-Lab/Mantis-8B-Idefics2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TIGER-Lab/Mantis-8B-Idefics2 with Transformers:

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

pipe = pipeline("image-text-to-text", model="TIGER-Lab/Mantis-8B-Idefics2")
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, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("TIGER-Lab/Mantis-8B-Idefics2")
model = AutoModelForImageTextToText.from_pretrained("TIGER-Lab/Mantis-8B-Idefics2")
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

vLLM

How to use TIGER-Lab/Mantis-8B-Idefics2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TIGER-Lab/Mantis-8B-Idefics2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TIGER-Lab/Mantis-8B-Idefics2",
		"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/TIGER-Lab/Mantis-8B-Idefics2

SGLang

How to use TIGER-Lab/Mantis-8B-Idefics2 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 "TIGER-Lab/Mantis-8B-Idefics2" \
    --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": "TIGER-Lab/Mantis-8B-Idefics2",
		"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 "TIGER-Lab/Mantis-8B-Idefics2" \
        --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": "TIGER-Lab/Mantis-8B-Idefics2",
		"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 TIGER-Lab/Mantis-8B-Idefics2 with Docker Model Runner:
```
docker model run hf.co/TIGER-Lab/Mantis-8B-Idefics2
```

No Chat Template

by dipta007 - opened Jul 15, 2024

Discussion

dipta007

Jul 15, 2024

I am getting this warning

No chat template is set for this processor, falling back to a default class-level template. This is very error-prone, because models are often trained with templates different from the class default! Default chat templates are a legacy feature and will be removed in Transformers v4.43, at which point any code depending on them will stop working. We recommend setting a valid chat template before then to ensure that this model continues working without issues.

Is it intentional?

DongfuJiang

TIGER-Lab org Jul 15, 2024

I think it's a warning due to the new features of transformers. It's not intentional by us. The code to run Mantis-8B-Idefics2 shall be the same as the original Idefics2. If you can get this warning running Mantis-8B-Idefics2, then I guess you can also get this warning running the original Idefics2.

dipta007

Jul 15, 2024

Not actually, there is no warning on the HuggingFaceM4/idefics2-8b model.
When I print the processor of mantis model, the chat template is null.
For immediate solution, I just loaded both of the processor and assigned the mantis one to the idefics2 one.

xkkk777

Aug 23, 2024

Not actually, there is no warning on the HuggingFaceM4/idefics2-8b model.
When I print the processor of mantis model, the chat template is null.
For immediate solution, I just loaded both of the processor and assigned the mantis one to the idefics2 one.

Yeah, I encounter the same issue here.

DongfuJiang

TIGER-Lab org Sep 2, 2024

I have updated the processor, and now it shall be working with

processor = AutoProcessor.from_pretrained("TIGER-Lab/Mantis-8B-Idefics2")

DongfuJiang changed discussion status to closed Sep 2, 2024

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