Text-to-World Bench
Can a short message become a completed real-world task? We tested text agents on reservations, food delivery, movie tickets, and flights.
Explore the benchmark7 agents11 tasks
Delphi / Research
Can a short message become a completed real-world task? We tested text agents on reservations, food delivery, movie tickets, and flights.
Explore the benchmarkWe give AI agents challenging tasks in math, coding, physics, and other domains to measure their propensity to cheat. We also study how teaching agents moral principles and frameworks influences their behavior.