METR

By METR

Independent evaluations of frontier models' autonomous capabilities.

Safety

WHAT IT DOES

Maker
METR
Website
metr.org
Category
Safety

An independent evaluator measuring the autonomous capabilities of frontier models. Their 50% time-horizon measurements, which say how long a software or ML-research task a model finishes on its own half the time, are what this site’s software forecast is fitted on.

Listed in the directory because it is part of the stack pushing toward AGI. Picked by the editors; none of it feeds a calculation on this site.

More of the stack

OpenAI · Assistant

ChatGPT

The mass-market on-ramp to frontier models: a general assistant for writing, reasoning and code.

Visit site

Anysphere · Coding

Cursor

AI-native code editor that edits across an entire repository.

Visit site

Perplexity AI · Research

Perplexity

Answer engine that cites its sources as it searches the live web.

Visit site

Missing
a tool?

Send it by email with a line on what it does. Every listing is reviewed by hand before it goes up.