List your company

Interactive environment task execution on AppWorld normal (test)

4,554Avg@8 SuccessCuES
-182.161,047.422,2773,506.58Oct 6, 2025Oct 15, 2025Oct 24, 2025Nov 3, 2025Nov 12, 2025Nov 21, 2025Dec 1, 2025

Evaluation results

MethodLinks
CuESParams=14B2025.124,5544,524——
Qwen2.5Params=32B2025.123,3163,473——
Qwen3Params=14B2025.123,0982,848——
Qwen3Params=32B2025.122,8793,212——
Qwen3Params=8B2025.121,7762,143——
Qwen2.5Params=14B2025.121,1761,429——
Qwen3Params=4B2025.12300325——
Qwen2.5Params=7B2025.12125187——
ReAct + ACEBase LLM=GPT-5.1, Adaptation mode=Online Adaptation, GT Labels=✗2025.1065.8—72.658.9
ReAct + ACEBase LLM=GPT-5.1, Adaptation mode=Offline Adaptation, GT Labels=✗2025.1061.3—67.355.4
ReAct + ACEBase LLM=GPT-5.1, Adaptation mode=Offline Adaptation, GT Labels=✓2025.1060.2—66.753.6
ReAct + DC (CU)Base LLM=GPT-5.1, Adaptation mode=Online Adaptation, GT Labels=✗2025.1058.9—62.555.4
ReAct + GEPABase LLM=GPT-5.1, Adaptation mode=Offline Adaptation, GT Labels=✓2025.1056.2—64.348.2
ReActBase LLM=GPT-5.1, Adaptation mode=None2025.1054.2—61.946.4
Qwen2.5Params=3B2025.12023——
BaselineMemory Configuration=Baseline2026.03——69.650
Trajectory-Informed Memory GenerationMemory Configuration=+Memory2026.03——73.264.3