Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Node features regression on METR-LA (test)

2.77MAE

RLSTG

2.69883.17943.664.1406Jan 20, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
2.77
2026.01
2.89
2026.01
2.9
2026.01
2.93
2026.01
2.98
2026.01
3.58
2026.01
3.78
2026.01
4.55