- Velvet
- Posts
- LLM evaluations: Experiments & Monitoring
LLM evaluations: Experiments & Monitoring
Run one-time and ongoing evaluations with Velvet
Today we’re excited to announce Velvet’s new evaluation features. LLMs are powerful, but inherently unpredictable. Run experiments to decide which models to use, and continuous monitoring to maintain confidence in AI features over time.
Experiments: Test which model is right for each use case.
Continuous monitoring: Ongoing tests on AI features in production.
Velvet’s evaluation features are easy to get set up, and fully customizable. Watch this quick demo video to learn how it works.
Try it out today at docs.usevelvet.com!
Run Evaluations with Velvet | Read the docs →