Benchmark family

AppWorld

A family groups the versions of one benchmark. It is not evaluated and has no applicable models of its own; each of its 2 definitions is resolved separately.

Interactive CodingStateful AgentUser In Loop AgentTool Use

Definitions

Compare versions →

Newest first, following the recorded successor chain. Each definition is resolved against models separately.

BenchmarkAppWorld
ActiveNewest in family
Scale
457 apis
Primary metric
Task Goal Completion
Applicable models
51
BenchmarkAppWorld-UL
Active
Derived from AppWorld
Scale
210 tasks
Primary metric
Task Goal Completion
Applicable models
51

Family methodology

No family-level methodology recorded.

Across definitions

Tasks: Multi App Interactive Task Execution, Infeasibility Handling, Requirement Clarification, User Confirmation Seeking, User In The Loop Tool Use
Sources: None recorded at family level
Last reviewed: 2026-09-23

Applicability does not mean evaluated; benchmark result/model score data is outside this UI contract.