Instructions to use HuggingFaceM4/Idefics3-8B-Llama3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HuggingFaceM4/Idefics3-8B-Llama3 with Transformers:

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

pipe = pipeline("image-text-to-text", model="HuggingFaceM4/Idefics3-8B-Llama3")
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("HuggingFaceM4/Idefics3-8B-Llama3")
model = AutoModelForMultimodalLM.from_pretrained("HuggingFaceM4/Idefics3-8B-Llama3")
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 HuggingFaceM4/Idefics3-8B-Llama3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HuggingFaceM4/Idefics3-8B-Llama3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HuggingFaceM4/Idefics3-8B-Llama3",
		"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/HuggingFaceM4/Idefics3-8B-Llama3

SGLang

How to use HuggingFaceM4/Idefics3-8B-Llama3 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 "HuggingFaceM4/Idefics3-8B-Llama3" \
    --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": "HuggingFaceM4/Idefics3-8B-Llama3",
		"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 "HuggingFaceM4/Idefics3-8B-Llama3" \
        --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": "HuggingFaceM4/Idefics3-8B-Llama3",
		"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 HuggingFaceM4/Idefics3-8B-Llama3 with Docker Model Runner:
```
docker model run hf.co/HuggingFaceM4/Idefics3-8B-Llama3
```

Text only generation for Idefics3

#22

by FreekCool - opened Oct 28, 2024

Discussion

FreekCool

Oct 28, 2024

Hi everyone, is it possible to only pass text to Idefics3, so text to text generation? I am using approximately the same code as with Idefics2, which seems to be in line with Idefics3. In the following code I do not pass images to the prompt:

Create inputs

        messages = [
            {
                "role": "user",
                "content": [
                ]
            }
        ]

        format_prompt = prompt.format(truncated_text)
        messages[0]['content'].append({"type": "text", "text": format_prompt})

        idefics_prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
        inputs = processor(text=idefics_prompt, return_tensors="pt")
        inputs = {k: v.to(DEVICE) for k, v in inputs.items()}

However, this line "inputs = processor(text=idefics_prompt, return_tensors="pt")" gives me the following error:

ValueError: The number of images in the text [0] and images [] should be the same.

Does anyone know if there is a solution? I installed the newest Transformers version also from source but it does not work. Thanks in advance!

eltorio

Nov 16, 2024

I have the same problem

Guanzheng

Nov 30, 2024

It works by just tokenizing (using Idefics3 tokenizer) the idefics_prompt as the input.

eltorio

Nov 30, 2024

It works by just tokenizing (using Idefics3 tokenizer) the idefics_prompt as the input.
@Guanzheng
Something like ?

inputs = tokenizer.apply_chat_template(
        messages,
        tokenize = True,
        add_generation_prompt = True, # False for fine-tuning
        return_tensors = "pt",).to(DEVICE)

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