research·IndependentNew

Agent Interp Envs

A framework for testing and evaluating AI agents across various task domains, designed for misalignment interpretability research.

About

A framework for testing and evaluating AI agents across various task domains, designed for misalignment interpretability research.

Advanced AI for research and writing

$20/mo

Elicit

AI research assistant for literature review

$12/mo

Consensus

AI search engine for scientific papers

$13/mo

Claude Opus 4

Anthropics most intelligent model for complex tasks

Usage-based pricing