Improve

Evaluations

Reproducible eval suites with deterministic, LLM-as-judge, and code-based scorers.

Loading evaluation suites…

Model comparison · support_qa.v4

18s ago

Prompt

Model

Accuracy

Cost / run

Latency

Cite

Loading comparison…