AppWorld-UL (Code Agent) - Confirmation Tasks: leaderboard
AppWorld-UL (2026) turns AppWorld test-challenge scenarios into user-in-the-loop tasks that are infeasible, underspecified or need confirmation, so an agent succeeds only if it also asks, informs or confirms with a simulated user. Interactive task goal completion in percent on the 102 confirmation-requiring tasks with a ReAct code agent; higher is better.
Metric: Interactive task goal completion (%; share of 102 tasks whose actions need explicit user confirmation first 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 February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.7 Max | 67.7 | #145 |
| 2 | GLM-5.1 | 66.7 | #179 |
| 3 | Claude Opus 4.7 | 60.8 | #79 |
| 4 | GPT-5.5 (Medium) | 58.8 | #49 (GPT-5.5) |
| 5 | Kimi K2.6 | 50 | #178 |
| 6 | DeepSeek V4 Pro | 48 | #188 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=appworld-ul-code-agent-confirmation-tasks · How It Works · Data refreshed daily, snapshot 2026-11-13.