Concordat Cooperative Airline - Claude Haiku 5.5 paired50: leaderboard

Concordat tests airline-service agents with cooperative and pressure-testing customers. Each result keeps the customer simulation and agent configuration separate. Task success (%).

Metric: Task success (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 2 models tracked.

Top models

#ModelScoreOverall rank
1Claude Haiku 5.5 (standard agent)88
2Claude Haiku 5.5 (Concordat loose confirmation)88

Interactive version: theaggregate.ai/benchmark?slug=concordat-cooperative-airline-claude-haiku-5-5-paired50 · How It Works · Data refreshed daily, snapshot 2026-10-29.