Trip+ - Soft Preferences: leaderboard
Trip+ (2026) has tool-using LLM agents plan and revise minute-level travel itineraries over multi-turn conversations with profiled travelers, changing requests and environment disruptions, in a fixed travel sandbox. This board is satisfaction of profile-derived preferences such as pace and comfort, averaged over the plans an agent produces when a plan is expected. Scores are on a 0 to 100 scale; higher is better.
Metric: Soft-preference satisfaction (0-100), a rate scaled by 100: profile-derived rule scores for pace, comfort, transport, budget and interests, averaged over the plans that pass the response-mode gate, on multi-turn personalized travel planning with traveler profiles, request changes and environment disruptions, the same OpenAI-compatible function-calling scaffold over a fixed travel sandbox for every model, temperature 0 where supported; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 27B (Non-reasoning) | 66.26 | #341 (Qwen 3.6 27B) |
| 2 | Gemini 3.1 Pro (Preview) | 63.89 | #94 |
| 3 | Kimi K2.6 | 63.8 | #178 |
| 4 | Qwen 3.5 27B (Non-reasoning) | 63.75 | #359 (Qwen 3.5 27B) |
| 5 | DeepSeek V4 Pro | 62.58 | #188 |
| 6 | GPT-5.4 Mini | 62.07 | #389 |
| 7 | Gemini 3 Flash (Preview) | 61.72 | #144 |
| 8 | GLM-4 32B | 61.41 | #1299 |
| 9 | GLM-5.1 | 61.36 | #179 |
| 10 | MiniMax-M2.7 | 60.76 | #518 |
| 11 | Seed 2.0 Pro | 60.31 | #211 |
| 12 | Hy3-preview | 59.57 | #316 |
| 13 | Gemma 4 26B A4B | 59.4 | #615 |
| 14 | GPT-5.4 | 58.54 | #123 |
| 15 | DeepSeek V3.2 | 58.42 | #406 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=trip-plus-soft-preferences · How It Works · Data refreshed daily, snapshot 2026-11-13.