[Bugfix] Fix ITL recording in serving benchmark (#7372)

This commit is contained in:
Roger Wang
2024-08-09 10:00:00 -07:00
committed by GitHub
parent 249b88228d
commit 70d268a399

View File

@@ -276,8 +276,9 @@ async def async_request_openai_completions(
output.ttft = ttft
# Decoding phase
output.itl.append(timestamp -
most_recent_timestamp)
else:
output.itl.append(timestamp -
most_recent_timestamp)
most_recent_timestamp = timestamp
generated_text += data["choices"][0]["text"]