OfficialSDKs

System One Adapter

By TypeSafe AI

Run System One-shaped evaluations through other model APIs to build comparisons with Jev.

Visit project

Implementation overview

TypeSafe's Python adapter preserves the typed evaluation interface while obtaining answers from OpenAI-compatible or Anthropic APIs. It includes synchronous and asynchronous clients plus usage and attempt diagnostics for examining comparison runs.

Before you start

The adapter calls other model providers; it is a comparison tool for Jev workflows.

Topics

evaluationcomparisonadapterPython

Request reference

Edit a small example and copy a real request before integrating the project.

Try a request