Deriva — Derive the Algorithm

← Projectsinference-api / L4Health + Readiness~16 min

one move: bound a failure

Summarize an inference workload with warmup exclusion, p95 latency, throughput, quality, and SLO verdicts. In the Inference API system, implement this level as a deterministic contract before composing it with the next service.

Required API

def benchmark_report(runs, quality_target, p95_limit_ms, throughput_target):
    return {'runs': ..., 'p95_ms': ..., 'throughput': ..., 'quality_ok': ..., 'slo_ok': ...}

Behavior

exclude warmup from measured runs
compute p95 and token throughput
require quality and performance SLOs together

Examples — tap to reveal

Constraints

Use only the Python standard library.
Return deterministic values so every run can be compared with the contract.
Do not mutate caller-owned inputs.

Opening…