Instructions to use tiiuae/falcon-40b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use tiiuae/falcon-40b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="tiiuae/falcon-40b", trust_remote_code=True)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("tiiuae/falcon-40b", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("tiiuae/falcon-40b", trust_remote_code=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use tiiuae/falcon-40b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "tiiuae/falcon-40b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "tiiuae/falcon-40b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/tiiuae/falcon-40b
- SGLang
How to use tiiuae/falcon-40b 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 "tiiuae/falcon-40b" \ --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": "tiiuae/falcon-40b", "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 "tiiuae/falcon-40b" \ --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": "tiiuae/falcon-40b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use tiiuae/falcon-40b with Docker Model Runner:
docker model run hf.co/tiiuae/falcon-40b
Run falcon on Mac
I tried to run the model on Mac Studio. It's said that MPS doesn't support bfloat16, so I changed the dtype in the sample code to float16. And I got the error:
loc("varianceEps"("(mpsFileLoc): /AppleInternal/Library/BuildRoots/ff32e6fb-db00-11ed-a068-428477786501/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShadersGraph/mpsgraph/MetalPerformanceShadersGraph/Core/Files/MPSGraphUtilities.mm":228:0)): error: input types 'tensor<1x61x1xf16>' and 'tensor<1xf32>' are not broadcast compatible
LLVM ERROR: Failed to infer result type(s).
What should I do to make it work?
I'm getting the same error on a different model
I success to run the model when I change bfloat16 to float32, it seems that the MPS doesn't support some dtype and some dtype doesn't compatible to each other.