Choose your path

Pick the Stratix entry point that fits what you're trying to do.

If you...	Start here
...want to know which model is best at X	Stratix Public
...want to compare two models head-to-head	Stratix Public — Compare two models
...want to run a private evaluation on your own data	Stratix Premium — Sign up → First evaluation
...want to build an LLM judge	Stratix Premium — First judge
...want to upload a trace and score it	Stratix Premium — First trace
...want to integrate evaluations into code or CI/CD	SDK quickstart
...are setting up for a team	Stratix Premium — Sign up and Personalized onboarding
...just want to test your AI quickly	Stratix Public — Test your AI

All three at once

Many teams use all three surfaces. A typical first day looks like:

1. Public — research the catalog, shortlist candidate models.
2. Premium — sign up, run a private evaluation against your data on the shortlist.
3. SDK — wire the winning evaluation into your code or CI gate.

You can switch between Public and Premium with the Switch toggle in the Premium top bar.

Common pitfalls

• Skipping the Public catalog and going straight to Premium. You'll waste ECU running candidates that public scores would have ruled out.
• Building a judge before you have a labeled set. Judges only work as well as the labels you tune them against. Even 30 labels makes a meaningful difference.
• Wiring CI gates before you've run the eval interactively. Start in the Premium UI; promote to CI once you trust the result.

Where to next

• Stratix Public quickstart
• Stratix Premium quickstart
• SDK quickstart