@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / e373853e12
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 22:00 UTC

[Frontend] Relax api url assertion for openai benchmarking (#6046)

James Whedbee committed Jul 1, 2024 at 18:39 UTC e373853e12c890964e21fa0ac9b46fee48fa7c76
1 file changed +4 -4
benchmarks/backend_request_func.py
+4 -4
@@ -225,8 +225,8 @@ async def async_request_openai_completions(
225 ) -> RequestFuncOutput:
226 api_url = request_func_input.api_url
227 assert api_url.endswith(
228 - "v1/completions"
229 - ), "OpenAI Completions API URL must end with 'v1/completions'."
228 + "completions"
229 + ), "OpenAI Completions API URL must end with 'completions'."
230
231 async with aiohttp.ClientSession(timeout=AIOHTTP_TIMEOUT) as session:
232 assert not request_func_input.use_beam_search
@@ -304,8 +304,8 @@ async def async_request_openai_chat_completions(
304 ) -> RequestFuncOutput:
305 api_url = request_func_input.api_url
306 assert api_url.endswith(
307 - "v1/chat/completions"
308 - ), "OpenAI Chat Completions API URL must end with 'v1/chat/completions'."
307 + "chat/completions"
308 + ), "OpenAI Chat Completions API URL must end with 'chat/completions'."
309
310 async with aiohttp.ClientSession(timeout=AIOHTTP_TIMEOUT) as session:
311 assert not request_func_input.use_beam_search