List your company

Navigation on FrozenLake 4x4

0.95RewardASK (32B)
-0.0380.21850.4750.7315Apr 2, 2026

Evaluation results

MethodLinks
ASK (32B)Model Size=32B, trained_on=8x82026.040.957.710046.17
ASK (72B)Model Size=72B, trained_on=8x82026.040.957.9910064.22
ASK (1.5B)Model Size=1.5B, trained_on=8x82026.040.376.9210042.98
ASK (14B)Model Size=14B, trained_on=8x82026.040.0756.5910052.95
PPOtrained_on=8x82026.04046.98——
ASK (0.5B)Model Size=0.5B, trained_on=8x82026.04041.1210061.71
ASK (3B)Model Size=3B, trained_on=8x82026.04044.5810064.52
ASK (7B)Model Size=7B, trained_on=8x82026.0406110095.57