Concurrency hotfix

This commit is contained in:
Alpay Ariyak
2023-12-28 19:03:46 -05:00
committed by GitHub
parent a247a3afe1
commit 4a6bf9f0ee
+1 -1
View File
@@ -8,7 +8,7 @@ serverless_config = ServerlessConfig()
llm, tokenizer = initialize_llm_engine()
def concurrency_modifier(current_concurrency) -> int:
return max(0, serverless_config.max_concurrency - current_concurrency)
return serverless_config.max_concurrency
async def handler(job: dict) -> Generator[dict, None, None]:
job_input = job["input"]