Instructions to use HuggingFaceM4/idefics2-8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HuggingFaceM4/idefics2-8b with Transformers:

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

pipe = pipeline("image-text-to-text", model="HuggingFaceM4/idefics2-8b")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("HuggingFaceM4/idefics2-8b")
model = AutoModelForImageTextToText.from_pretrained("HuggingFaceM4/idefics2-8b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use HuggingFaceM4/idefics2-8b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HuggingFaceM4/idefics2-8b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HuggingFaceM4/idefics2-8b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/HuggingFaceM4/idefics2-8b

SGLang

How to use HuggingFaceM4/idefics2-8b 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/idefics2-8b" \
    --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": "HuggingFaceM4/idefics2-8b",
		"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 "HuggingFaceM4/idefics2-8b" \
        --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": "HuggingFaceM4/idefics2-8b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use HuggingFaceM4/idefics2-8b with Docker Model Runner:
```
docker model run hf.co/HuggingFaceM4/idefics2-8b
```

shape mismatch: value tensor of shape [2320] cannot be broadcast to indexing result of shape [2262]

#52

by yeargun - opened May 16, 2024

Discussion

yeargun

May 16, 2024

I am trying the very first example given in the documentation, yet I'm kind of new to deploying LLMs and stuff.

processor = AutoProcessor.from_pretrained("HuggingFaceM4/idefics2-8b")
model = AutoModelForVision2Seq.from_pretrained(
    "HuggingFaceM4/idefics2-8b",
    # torch_dtype=torch.float16,
     # _attn_implementation="flash_attention_2",
    # quantization_config=quantization_config,
)

prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
inputs = processor(text=prompt, images=[image1, image2], return_tensors="pt")

generated_ids = model.generate(**inputs, max_new_tokens=1500)
generated_texts = processor.batch_decode(generated_ids, skip_special_tokens=True)

And getting the following error
shape mismatch: value tensor of shape [2320] cannot be broadcast to indexing result of shape [2262]

VictorSanh

May 17, 2024

hi @yeargun
do you happen to have a full traceback?

sipie800

May 22, 2024

same

VictorSanh

May 22, 2024

same question @sipie800 , can you share a full traceback?

sipie800

May 24, 2024

by me the problem has been fixed somehow, but I can't recall it. It might be the transformers version issue.

chenduo

Jun 7, 2024

same issue, anyone can help me to solve this problem?😭

mdmev

Jun 8, 2024

I have a similar issue; I posted it on the chatty version. In case it helps, issue

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