Instructions to use facebook/chameleon-30b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/chameleon-30b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="facebook/chameleon-30b")# Load model directly from transformers import AutoProcessor, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("facebook/chameleon-30b") model = AutoModelForImageTextToText.from_pretrained("facebook/chameleon-30b") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use facebook/chameleon-30b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "facebook/chameleon-30b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "facebook/chameleon-30b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/facebook/chameleon-30b
- SGLang
How to use facebook/chameleon-30b 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 "facebook/chameleon-30b" \ --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": "facebook/chameleon-30b", "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 "facebook/chameleon-30b" \ --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": "facebook/chameleon-30b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use facebook/chameleon-30b with Docker Model Runner:
docker model run hf.co/facebook/chameleon-30b
Error: "Transformers does not recognize this architecture."
Getting this error when trying to run in OobaBooga: "ValueError: The checkpoint you are trying to load has model type chameleon but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date."
88GB of VRAM. Updated Transformers via pip and still get the same error.
The inference code isn't merged yet to transformers' git: huggingface/transformers#31534. There is still some work before it gets merged into git.
Inference code is merged, the next Transformers PyPI release will have compatibility, it's not added as of 4.42.4