GroupTravelBench - Group Fairness: leaderboard
Metric: Group Fairness (%): 100 x the lowest over the highest per-user utility of the final itinerary (the released code scores 0 when utilities straddle zero and keeps every task within 0 to 100), averaged over all 650 tasks, on GroupTravelBench, multi-user multi-turn group travel planning in a group-chat sandbox with 10 cached travel tools (DeepSeek-V4-Flash user simulator at temperature 0, agent temperature 0.7, mean of 3 trials per task); GPT-5.1 in instruct mode, the other models with reasoning enabled; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.6 Max (Preview) (Thinking) | 52.1 |
| 2 | GPT-5.1 (Non-reasoning) | 48.1 |
| 3 | Qwen 3.5 Plus (Thinking) | 44.9 |
| 4 | Qwen 3.5 Flash (Thinking) | 41.9 |
| 5 | Qwen 3 4B Instruct | 41.7 |
Interactive version: theaggregate.ai/benchmark?slug=grouptravelbench-group-fairness · How It Works · Data refreshed daily, snapshot 2026-10-07.