Loading...
Loading...
Contributors:
Dataset URI:
10.5281/zenodo.23605512Stressor suites, labeled contaminated workloads, and per-request reference runs for stress testing open-weight LLM deployments.D1, stressor suites: 883 pairs of an ordinary prompt and its LoopLLM stressor in 12 suites for 12 open-weight checkpoints. Four suites (646 pairs) are validated during optimization (Stage A) and on a vLLM deployment under load (Stage B); eight generated suites carry Stage A only. Three control-stressor sets.D2, labeled workloads: replayable CSV workloads at contamination ratios 0 to 0.15 with manifests and configurations.D3, reference runs: 24 runs on four capacity-calibrated single-GPU vLLM deployments, 796,500 request rows, SLO budgets, vLLM scheduler telemetry, and two GuideLLM runs.The record also contains a source snapshot of the contambench Python package (Apache-2.0), which fetches this record, reproduces the published tables, generates suites for new models, and runs campaigns against an endpoint you operate. Start with README.md and DATASET_CARD.md.The stressors raise serving cost by design. Run them only against systems you operate or are authorized to test.
Keyword(s):
dataset; GuideLLM; LLM serving; service-level objectives; stress testing.
MeSH Term(s):
Grants and Funding:
No Grants or Funding Data Available
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)