@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 96aa014d1e
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 17, 03:00 UTC

fix benchmark format reporting in buildkite (#3693)

Simon Mo committed Mar 28, 2024 at 14:35 UTC 96aa014d1e299bede81b30d6bb7844739f68b849
1 file changed +3 -1
.buildkite/run-benchmarks.sh
+3 -1
@@ -49,7 +49,9 @@ sed -n '$p' benchmark_throughput.txt >> benchmark_results.md # last line
49 echo "### Serving Benchmarks" >> benchmark_results.md
50 sed -n '1p' benchmark_serving.txt >> benchmark_results.md # first line
51 echo "" >> benchmark_results.md
52 -tail -n 13 benchmark_serving.txt >> benchmark_results.md # last 13 lines
52 +echo '```' >> benchmark_results.md
53 +tail -n 20 benchmark_serving.txt >> benchmark_results.md # last 20 lines
54 +echo '```' >> benchmark_results.md
55
56 # upload the results to buildkite
57 /workspace/buildkite-agent annotate --style "info" --context "benchmark-results" < benchmark_results.md