UserToolBench - Lack of Information: leaderboard

Metric: Exact trajectory accuracy (%; lack-of-information tasks, where the model must ask for clarification or infer the missing constraint from prior behavior; a trajectory counts when it matches the profile-conditioned reference: the tool name and decision-relevant arguments, the ordered call sequence for multi-tool tasks, or the correct clarify-or-infer decision when information is missing; 799 profile-hidden evaluation instances from long-horizon interaction trajectories of 10 privacy-sanitized user profiles over 36 tool sets; the model sees the interaction history, the current request and the tool schemas but not the profile). Source: arxiv.org. Saturation forecast: Around 2030. 9 models tracked.

Top models

#ModelScore
1GPT-5.442.22
2GLM-539.07
3Kimi K2.637.2
4DeepSeek V4 Pro32.29
5Gemini 3.5 Flash19.99
6Qwen 3.6 Plus19.52

Interactive version: theaggregate.ai/benchmark?slug=usertoolbench-lack-of-information · How It Works · Data refreshed daily, snapshot 2026-09-29.