AIPerf

InferCrane delegates load generation to AIPerf and persists the workload, runtime, model artifact, provider, GPU, revision, results, and exact reproduction command.

infercrane benchmark qwen-prod \
  --requests 1000 \
  --concurrency 32 \
  --random-seed 17

Use --revision candidate only when you intend to benchmark isolated candidate capacity. Benchmark data remains in the InferCrane control plane; it is not uploaded by default.

Benchmarking

Understand measurements, persisted reproduction metadata, and evidence limits.

⌘I