List your company

Agentic Planning and Ideation on MUTATE (test)

41Success Count (Path Discovery)ReDNA
11.8819.442734.56May 27, 2026

Evaluation results

MethodLinks
ReDNAModel=Claude Sonnet 4.62026.054132.82.6163.2053.533
ReDNAModel=GPT-5.42026.053840.22.7713.2623.725
EscapeAgentModel=Claude Sonnet 4.62026.053743.92.5262.7433.341
ReDNAModel=Qwen3 235B A22B2026.053636.52.7073.4073.353
EscapeAgentModel=GPT-5.42026.053544.92.5422.993.516
EscapeAgentModel=Qwen3 235B A22B2026.053448.32.6453.1483.47
BaseModel=Claude Sonnet 4.62026.053026.92.7643.0793.35
Self-RefineModel=GPT-5.42026.0529332.7382.7033.514
ReDNAModel=Llama 4 Maverick2026.052945.32.4093.0213.031
BaseModel=GPT-5.42026.052726.32.6472.8273.588
Self-RefineModel=Qwen3 235B A22B2026.052735.72.7243.1813.449
Self-RefineModel=Claude Sonnet 4.62026.052521.92.7333.0883.457
EscapeAgentModel=Llama 4 Maverick2026.052352.72.2142.5713.11
BaseModel=Qwen3 235B A22B2026.052233.82.7553.1153.388
BaseModel=Llama 4 Maverick2026.051336.22.2562.6673.128
Self-RefineModel=Llama 4 Maverick2026.051331.22.3242.6813.181