Share your thoughts, 1 month free Claude Pro on usSee more

Agentic Task Solving on AB-DB

51.1Pass@3

A³

Updated 23d ago

Evaluation Results

Method	Links
A³ 2026.05		51.1	55.3
A³ 2026.05		50.3	55
GSPO 2026.05		42.8	45.8
RetroAgent 2026.05		41.8	44.7
ReACT 2026.05		38.2	41.2
HGPO 2026.05		37.8	41.2
LATS 2026.05		37.7	41.2
rStar 2026.05		36.1	41.2
GiGPO 2026.05		35.8	38.9