List your company

Agentic Task Completion on AppWorld (test-normal)

56.5AccuracyACON UTCO
10.7422.6234.546.38Oct 1, 2025

Evaluation results

MethodLinks
ACON UTCOAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1056.522.827.334.69
ACON UTCOAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1056.2—7.484.43
No compressionAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=None2025.105616.149.935.96
ACON UTCOAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.1053.618.127.434.93
ACON UTAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1051.220.927.174.49
ACON UTAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1050—7.394.47
No compressionAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=None2025.1047.9—10.15.36
ACON UTCOAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.1047.9—7.34.43
ACON UTAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.104716.677.625.08
FIFOAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1045.828.486.735.69
ACON UTAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.1045.8—7.795
PromptingAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1043.524.016.935.29
PromptingAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.1042.317.386.584.09
PromptingAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1041.7—7.15.36
FIFOAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1039.6—7.266.24
LLMLinguaAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1039.324.427.56.37
LLMLinguaAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1037.5—8.047.39
PromptingAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.1035.4—6.964.11
LLMLinguaAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.1032.118.168.176.01
LLMLinguaAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=Observation Compression2025.1029.2—8.245.23
RetrievalAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1027.433.178.396.68
RetrievalAgent=gpt-4.1, Compressor=gpt-4.1, Compression Type=History Compression2025.1012.5—8.747.72