Optimize Triton MoE Kernel (#2979)

Co-authored-by: Cade Daniel <edacih@gmail.com>
This commit is contained in:
Philipp Moritz
2024-02-26 13:48:56 -08:00
committed by GitHub
parent 70f3e8e3a1
commit cfc15a1031
7 changed files with 296 additions and 14 deletions

View File

@ -432,7 +432,9 @@ def get_requirements() -> List[str]:
return requirements
package_data = {"vllm": ["py.typed"]}
package_data = {
"vllm": ["py.typed", "model_executor/layers/fused_moe/configs/*.json"]
}
if os.environ.get("VLLM_USE_PRECOMPILED"):
ext_modules = []
package_data["vllm"].append("*.so")