Enabling cache for torch.compile

Enabling cache for torch.compile#

SGLang uses max-autotune-no-cudagraphs mode of torch.compile. The auto-tuning can be slow. If you want to deploy a model on many different machines, you can ship the torch.compile cache to these machines and skip the compilation steps.

This is based on https://pytorch.org/tutorials/recipes/torch_compile_caching_tutorial.html

  1. Generate the cache by setting TORCHINDUCTOR_CACHE_DIR and running the model once.

TORCHINDUCTOR_CACHE_DIR=/root/inductor_root_cache python3 -m sglang.launch_server --model meta-llama/Llama-3.1-8B-Instruct --enable-torch-compile

  1. Copy the cache folder to other machines and launch the server with TORCHINDUCTOR_CACHE_DIR.