AppWorld-UL (Code Agent) - Underspecified Tasks: leaderboard

Metric: Interactive task goal completion (%; share of 102 tasks with a required detail withheld, which the agent must ask for whose AppWorld unit tests all pass, including a test that the agent asked the simulated user for every piece of hidden information it needed; ReAct code agent that writes Python against the AppWorld app APIs; GPT-5.5 user simulator answering from a fixed knowledge set). Source: arxiv.org. Saturation forecast: Around December 2027. 6 models tracked.

Top models

#ModelScore
1Claude Opus 4.750
2GPT-5.5 (Medium)38.2
3GLM-5.136.3
4Qwen 3.7 Max33.3
5Kimi K2.628.4
6DeepSeek V4 Pro25.5

Interactive version: theaggregate.ai/benchmark?slug=appworld-ul-code-agent-underspecified-tasks · How It Works · Data refreshed daily, snapshot 2026-09-29.