scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-12 03:48:56 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 57

run_benchmark_rps_suite: default 100k requests per run

Browse source
This commit is contained in:
Alexsander Hamir 2026-02-02 16:02:35 -08:00
parent c2073bdf1c
commit cc0dfc8949
1 changed files with 6 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
scripts/run_benchmark_rps_suite.py
View file

@ -4,7 +4,7 @@ Run the proxy-vs-provider benchmark 30 times: 1k, 2k, ..., 30k RPS, one at a tim
Each run writes to a separate results file (e.g. benchmark_results_1000rps.txt, ...).
Usage:
# Wait for proxy, then run 30 tests (1k–30k RPS), 1M requests each:
# Wait for proxy, then run 30 tests (1k–30k RPS), 100k requests each:
python scripts/run_benchmark_rps_suite.py
# Skip proxy wait (proxy already up):
@ -22,13 +22,13 @@ import os
import subprocess
import sys
# RPS values for the runs: 1k, 2k, ..., 30k
RPS_LEVELS = list(range(1000, 31_000, 1000))
# RPS values for the runs: 1k, 2k, ..., 10k
RPS_LEVELS = list(range(1000, 11_000, 1000))
def main() -> int:
parser = argparse.ArgumentParser(
description="Run benchmark 30 times at 1k–30k RPS, one at a time, each to a separate file.",
description="Run benchmark 10 times at 1k–10k RPS, one at a time, each to a separate file.",
formatter_class=argparse.RawDescriptionHelpFormatter,
epilog=__doc__,
)
@ -51,8 +51,8 @@ def main() -> int:
parser.add_argument(
"--requests",
type=int,
default=1_000_000,
help="Number of requests per run (default: 1M)",
default=20_000,
help="Number of requests per run (default: 20k)",
)
parser.add_argument(
"--no-wait",