The AI arena is free today

Open Superagent
AI OrganizationMaintains 9 benchmarks

Laude Institute: AI benchmarks & research

Non-profit lab that runs Terminal-Bench with Stanford, plus the Terminus reference agent.

Maintained by Laude Institute

9 benchmarks

In one line

Laude Institute maintains Terminal-Bench (counts toward the LLM Stats Score, 25 models scored), Terminal-Bench 2.0 (counts toward the LLM Stats Score, 53 models scored), Terminal-Bench 2.1 (counts toward the LLM Stats Score, 56 models scored), Terminal-Bench 3.0 (counts toward the LLM Stats Score, 4 models scored), Terminal-Bench 4.0 (counts toward the LLM Stats Score, 37 models scored), Terminal-Bench Hard (counts toward the LLM Stats Score, 2 models scored), Terminal-Bench-Science 0.1 (counts toward the LLM Stats Score, 13 models scored), Terminus (counts toward the LLM Stats Score, 1 model scored) and Terminal-Bench 4.0-AA (counts toward the LLM Stats Score, 5 models scored). It ships no models of its own on LLM Stats.

FAQ

Common questions about Laude Institute.

What is Laude Institute?

Non-profit lab that runs Terminal-Bench with Stanford, plus the Terminus reference agent. On LLM Stats it maintains 9 benchmarks, led by Terminal-Bench 2.1, Terminal-Bench 2.0 and Terminal-Bench 4.0. It ships no models of its own; each score on its benchmarks carries its own provenance label on the model page.

Which Laude Institute benchmarks count toward the LLM Stats Score?

Terminal-Bench 2.1, Terminal-Bench 2.0, Terminal-Bench 4.0, Terminal-Bench, Terminal-Bench-Science 0.1, Terminal-Bench 4.0-AA and 3 more count toward the LLM Stats Score. Only benchmarks admitted to the core set move the main leaderboard; the rest are shown for transparency.

Which models lead Laude Institute's benchmarks?

SWE-2 leads Terminal-Bench 2.1 at 92.8% (56 models scored), GPT-5.5 leads Terminal-Bench 2.0 at 82.7% (53 models scored) and Claude Sonnet 5.5 leads Terminal-Bench 4.0 at 70.6% (37 models scored). Each benchmark page lists the full ranking and how the results were produced.

Where are Laude Institute's official benchmark results?

Laude Institute's official site is https://www.tbench.ai. Each benchmark page on LLM Stats links to that benchmark's own paper or leaderboard, and shows the scores next to every other benchmark so models can be compared on one page.