Benchmark RadarAI BENCHMARK PROFILE
GroupTravelBench
A benchmark for multi-user, multi-turn travel planning with 650 tasks and a synchronous group-chat sandbox, evaluating elicitation, coordination, and fairness-aware planning.
- Released
- 2026-05-24
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Highlights the challenge of group-level outcome quality for LLM agents.
Motivation
Travel planning in the real world is overwhelmingly a \textit{group} activity, yet existing LLM travel-planning benchmarks reduce it to a single user, where the field is approaching saturation.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.