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