History

Nick Hill d4db9f53c8 [Benchmark] Add `--async-engine` option to benchmark_throughput.py (#7964 )		2024-09-03 20:57:41 -04:00
..
cutlass_benchmarks	[Kernel] Add per-tensor and per-token AZP epilogues (#5941 )	2024-08-06 18:17:08 +00:00
kernels	[Kernel] Replaced `blockReduce[...]` functions with `cub::BlockReduce` (#7233 )	2024-08-21 20:18:00 -04:00
overheads	[Frontend] Add FlexibleArgumentParser to support both underscore and dash in names (#5718 )	2024-06-20 17:00:13 -06:00
backend_request_func.py	[misc] Add Torch profiler support (#7451 )	2024-08-21 15:39:26 -07:00
benchmark_latency.py	[Frontend] Refactor prompt processing (#4028 )	2024-07-22 10:13:53 -07:00
benchmark_prefix_caching.py	[Misc] Enhance prefix-caching benchmark tool (#6568 )	2024-08-22 09:32:02 -07:00
benchmark_serving.py	Add more percentiles and latencies (#7759 )	2024-08-29 16:48:11 -07:00
benchmark_throughput.py	[Benchmark] Add `--async-engine` option to benchmark_throughput.py (#7964 )	2024-09-03 20:57:41 -04:00
launch_tgi_server.sh	[benchmark] Update TGI version (#7917 )	2024-08-27 15:07:53 -07:00
README.md	Change the name to vLLM (#150 )	2023-06-17 03:07:40 -07:00
sonnet.txt	feat(benchmarks): Add Prefix Caching Benchmark to Serving Benchmark (#3277 )	2024-03-27 13:39:26 -07:00

Benchmarking vLLM

Downloading the ShareGPT dataset

You can download the dataset by running:

wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/resolve/main/ShareGPT_V3_unfiltered_cleaned_split.json