Benchmarks
Alerts Guide
How to configure alert thresholds, what each alert type means, and how to act on score_drop, cost_spike, latency_spike, and model_version_change alerts.
This guide is coming soon. In the meantime, see the Benchmark Guide for how to set up benchmark suites, and the Understanding Drift guide for how anomaly detection works.
Get alerted before users notice
Pro and above. Set thresholds on any benchmark and receive alerts the moment a run breaches them.
See pricingWhat’s next
Benchmark Guide
Save a run configuration, schedule it on a cadence, and track quality and cost drift over time.
Understanding Drift
What model drift is, how LLM Prover detects it, and how to act on anomaly flags.
Diagnostics Guide
How the diagnostic judge works, what it flags, and how to act on findings in comparisons, evaluations, and benchmarks.
Evaluations Guide
Score model outputs against a rubric or gold standard answer and get per-criterion reasoning.