We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent a7b4ef3 commit 7805b5dCopy full SHA for 7805b5d
1 file changed
benchmarks/benchmark_serving.py
@@ -338,6 +338,12 @@ def calculate_metrics(
338
)
339
total_output += output_len
340
total_input += input_requests[i].prompt_len
341
+ if output_len == 0:
342
+ print(
343
+ f"""-------- output_len is zero for {i}th request:,
344
+ output: {outputs[i]}"""
345
+ )
346
+ continue
347
per_token_latencies.append(outputs[i].latency / output_len)
348
ttfts.append(outputs[i].ttft)
349
completed += 1
0 commit comments