AppWorld-UL (Code Agent) - Scenario Goal Completion: leaderboard

Metric: Interactive scenario goal completion (%; share of task scenarios in which the agent completes every task variant under the same interaction checks; 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 April 2028. 6 models tracked.

Top models

#ModelScore
1Claude Opus 4.730.2
2GPT-5.5 (Medium)23.5
3Qwen 3.7 Max22.9
4GLM-5.121.2
5Kimi K2.612.5
6DeepSeek V4 Pro7.3

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