Instructions to use amazon/FalconLite with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use amazon/FalconLite with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="amazon/FalconLite", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("amazon/FalconLite", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use amazon/FalconLite with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "amazon/FalconLite" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "amazon/FalconLite", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/amazon/FalconLite
- SGLang
How to use amazon/FalconLite 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 "amazon/FalconLite" \ --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": "amazon/FalconLite", "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 "amazon/FalconLite" \ --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": "amazon/FalconLite", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use amazon/FalconLite with Docker Model Runner:
docker model run hf.co/amazon/FalconLite
langchain
Does it integrate well with langchain? Are there examples?
Since we host this model using Huggingface text generation inference, to the best of my knowledge, you can refer to https://python.langchain.com/docs/integrations/llms/huggingface_textgen_inference to see the example how to use it in langchain. Cheers!
Would love a tutorial on how to set this up for hosting and creating an api endpoint for querying via http on aws. or make it an inference endpoint deployable on huggingface (please also on eu aws computing centers, for DSGVO and DPA approval). Cheers!
FYI: @dm-mschubert https://github.com/awslabs/extending-the-context-length-of-open-source-llms/pull/3 We are working on it now :)
Hi @dm-mschubert , we have a notebook to deploy FalconLite onto a SageMaker endpoint running on AWS - https://github.com/awslabs/extending-the-context-length-of-open-source-llms/blob/main/custom-tgi-ecr/deploy.ipynb
Feel free to give it a try and let us know if any issues. Thanks
Hi @dm-mschubert , we have a notebook to deploy FalconLite onto a SageMaker endpoint running on AWS - https://github.com/awslabs/extending-the-context-length-of-open-source-llms/blob/main/custom-tgi-ecr/deploy.ipynb
Feel free to give it a try and let us know if any issues. Thanks
Which SageMaker Image and Kernel should we use for https://github.com/awslabs/extending-the-context-length-of-open-source-llms/blob/main/custom-tgi-ecr/deploy.ipynb?