List your company

Sequential task management and state maintenance on Lifelong AgentBench

100AccuracyGA
68.876.98593.1Apr 18, 2026Apr 26, 2026May 4, 2026May 12, 2026May 20, 2026May 28, 2026Jun 6, 2026

Evaluation results

MethodLinks
GAModel=Claude Sonnet 4.62026.04100222202414.15
GAModel=Claude Sonnet 4.62026.06100222202414.15
BA-IncModel=deepseek-v4-flash2026.061007868423.85
GAModel=deepseek-v4-pro2026.061003853441847.82
BA-FullModel=deepseek-v4-pro2026.061004053543945.53
BA-IncModel=deepseek-v4-pro2026.06100000—
GAModel=Minimax M2.72026.0490400234232.12
GAModel=deepseek-v4-flash2026.06906494269026.07
BA-FullModel=deepseek-v4-flash2026.06856344067425.23
Claude CodeModel=Claude Sonnet 4.62026.0475800148140.92
Claude CodeModel=Claude Sonnet 4.62026.0675800148140.92
OpenClawModel=Claude Sonnet 4.62026.04701.43211.450.48
OpenClawModel=Minimax M2.72026.04701.2171.220.57
OpenClawModel=Claude Sonnet 4.62026.06701.43211.450.48