Sglang Auto Benchmark

Run SGLang auto benchmark searches with tiered server-flag sweeps, canonical dataset preparation, ShareGPT auto-download, custom-data conversion/validation, SLA or fixed-QPS benchmarking, CSV export, and optional second-stage speculative/EAGLE tuning. Use when the user wants an AI-operated benchmark workflow rather than a one-off bench_serving command.

futuremls-lab Updated
OSCAR pipeline

File contents

futuremls-lab/oscar commit 21dd803208

Frequently asked questions

npx get-skill add futuremls-lab/sglang-auto-benchmark