Skip to content
AI providers & safety

Multiple model providers, one learning standard.

Wizzist can use model APIs from Anthropic, Google, and OpenAI. The provider supplies the model; Wizzist supplies the learning workflow, relevant context, and tutoring instructions.

Configured model providers.

Wizzist uses provider APIs rather than the providers' consumer chat products. Configured model versions can change as availability, quality, and cost change.

How routing works

Predictable routing with provider fallback.

Each Wizzist feature has a configured model choice. Account or Team preferences can influence eligible requests, and the router can try another configured provider when the first one is unavailable.

Task-specific configuration

Tutoring, lesson creation, grading, planning, and other features can have different model settings instead of sending every request to one generic chat engine.

Relevant learner context

A workflow can include the goals, recent work, or learning signals needed for that request. Wizzist does not promise that every provider receives every part of a profile.

Fallbacks, not magic

Routing follows configured provider order, preferences, and budget constraints. It is not a real-time contest that silently picks a perfect model for every prompt.

Responsible use

Learning rules and honest limits.

Model output can be incomplete or wrong. Wizzist adds tutoring instructions and product controls, but learners should still check important facts and follow school rules.

  • Tutor workflows ask for learner effort and use hints before stronger help.
  • Provider requests should include only the context needed for the current workflow.
  • Private tutor sessions are not presented as an automatic adult transcript feed.
  • Model and feature activity supports product operation, safety, and usage accounting.
  • Wizzist does not sell personal information or use student work for advertising.

Read the Wizzist Privacy Policy

Questions about Wizzist AI?

Ask about a workflow, provider, or school use case.

We will tell you what is available now and what still needs product work.