Skip to content
Open
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
fix bench script
Signed-off-by: jthomson04 <[email protected]>
  • Loading branch information
jthomson04 committed Dec 9, 2025
commit 71a88ea36bcefbf99d3d38f8be8eb8431c2eff57
20 changes: 0 additions & 20 deletions components/backends/trtllm/performance_sweeps/scripts/bench.sh
Original file line number Diff line number Diff line change
Expand Up @@ -232,26 +232,6 @@ for concurrency in ${concurrency_list}; do
-H 'Authorization: Bearer NOT USED' \
-H 'Accept: text/event-stream'
fi
python3 ${SCRIPTS_DIR}/scripts/bench/benchmark_serving.py \
--served-model-name ${model} \
--model ${model_path} \
--dataset-name random \
--num-prompts "$num_prompts" \
--random-input-len ${isl} \
--random-output-len ${osl} \
--random-range-ratio 0.8 \
--use-chat-template \
--ignore-eos \
--use-chat-template \
--backend "dynamo" \
--endpoint "/v1/completions" \
--percentile-metrics ttft,tpot,itl,e2el \
--max-concurrency "$concurrency" \
--host ${hostname} \
--port ${port} \
--save-result \
--result-dir "${log_path}/results" \
--result-filename "results_concurrency_${original_concurrency}_gpus_${total_gpus}_ctx_${prefill_gpus}_gen_${decode_gpus}.json"

echo "Benchmark with concurrency ${concurrency} done"
done
Expand Down
Loading