Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

AIRS-Bench — leaderboard

AI research science benchmark for autonomous ML research agents, built from tasks that require experimentation, coding, and scientific judgment.

Source: github.com.

Interactive version: theaggregate.ai/benchmark?slug=airs-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.