evaluate( )
Pattern
Jev Evals
Run a rubric eval suite with jev-evals: one Jev call per case answers every rubric, and a baseline diff catches fractional regressions.
IntermediateSep 2026View the full pattern on desktop
Problems solved
- Judge every rubric for a case in one call instead of one call per rubric
- Show what an eval run actually cost from real usage
- Catch fractional regressions against a baseline run
Use cases
- Per-PR eval suites for agents
- Baseline vs candidate regression checks
- Teaching one-round-trip rubric judging
AI SDK APIs
experimental_evaluate
Tags
jevjev-evalsevaluationllm-as-judgerubricregressionai-sdk
Requirements
- Node.js
- 24+
- Memory
- 256MB
- Other
- Vercel AI Gateway API key
Environment variables
AI_GATEWAY_API_KEYGet key
External services
vercel-ai-gatewayupstash
Ready to use?
View the full interactive pattern with live preview and copy the source code.
View the pattern on desktop
