List your company

Agentic Tool-use on AppWorld (Challenge)

83.7TGCRCL (all primitives)
61.75667.45373.1578.847Apr 3, 2026

Evaluation results

MethodLinks
RCL (all primitives)Agent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0483.7
ACE + BatchingAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0483.4
ACE + Failure ReplayAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0481.8
ACE + Grouped RolloutsAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0481.4
ACEAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0480.7
ACE + Auxiliary LossesAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0480.6
ACE + Optimizer StateAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0479.4
ACE + Credit AssignmentAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0479.4
SeedAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0475.8
ACE + Auxiliary LossesAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0474.6
ACE + Optimizer StateAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0473.9
ACE + Failure ReplayAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0473.1
ACE + Grouped RolloutsAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0472.4
RCL (all primitives)Agent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0471.9
ACE + Credit AssignmentAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0471.5
ACE + BatchingAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0471.2
ACEAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0469.1
GEPAAgent Model=GPT-5.4 Nano, Optimizer Model=Claude Opus, Optimization Iterations=302026.0466.7
SeedAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0464.3
GEPAAgent Model=Gemini 3.1 Flash Lite, Optimizer Model=Claude Opus, Optimization Iterations=302026.0462.6