ai-workspace-services/litellm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: add auth headers and empty latencies guard to benchmark script

Browse Source
This commit is contained in:
Ryan Crabbe 2026-02-23 09:58:40 -08:00
parent 8244ad1f0e
commit 95d9514054
1 changed files with 14 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
scripts/benchmark_mock.py
View File

@ -16,12 +16,17 @@ REQUEST_BODY = {
"user": "new_user", "user": "new_user",
} }
HEADERS = {
"Authorization": "Bearer sk-1234",
"Content-Type": "application/json",
}
async def send_request(session, url, semaphore): async def send_request(session, url, semaphore):
async with semaphore: async with semaphore:
start = time.perf_counter() start = time.perf_counter()
try: try:
async with session.post(url, json=REQUEST_BODY) as resp: async with session.post(url, json=REQUEST_BODY, headers=HEADERS) as resp:
await resp.read() await resp.read()
elapsed = time.perf_counter() - start elapsed = time.perf_counter() - start
return elapsed if resp.status == 200 else None return elapsed if resp.status == 200 else None
@ -50,6 +55,14 @@ async def run_benchmark(url, n_requests, max_concurrent):
latencies = [r for r in results if r is not None] latencies = [r for r in results if r is not None]
failures = sum(1 for r in results if r is None) failures = sum(1 for r in results if r is None)
if not latencies:
return {
"mean": 0, "p50": 0, "p95": 0, "p99": 0,
"throughput": 0, "failures": n_requests,
"wall_time": wall_elapsed, "n_requests": n_requests,
"max_concurrent": max_concurrent, "latencies": [],
}
latencies.sort() latencies.sort()
n = len(latencies) n = len(latencies)
mean = statistics.mean(latencies) * 1000 mean = statistics.mean(latencies) * 1000