PersonalAgentEval

The best personal agent adapts to different users and preferences across domains and contexts.

MatrAIx Core Team

Leaderboard

Leaderboard

Efficiency and Cost

The Simulated User

A language model plays each person; it never sees their accounts or the grader. Replay a run:

Resolved Rate by Capability

Profiles

Average of all personal agents

By Domain
By Capability

Want your personal agent evaluated? Contact us.