List your company

Language Modeling on WikiText (PPL, Util. (TFLOP/s), Latency, Throughput)

17.69PPLVanilla-16L
17.644817.949918.25518.5601Jun 25, 2026

Evaluation results

MethodLinks
Vanilla-16LParams (B)=1.28, Cost (TFLOPs)=4.822026.0617.6953.8789.4722,890
CascadeFormer-A1Params (B)=1.25, Cost (TFLOPs)=4.722026.0617.7955.3985.1724,048
Vanilla-15LParams (B)=1.21, Cost (TFLOPs)=4.542026.0617.8454.2383.8124,439
CascadeFormer-A2Params (B)=1.22, Cost (TFLOPs)=4.582026.0617.8459.7776.6226,731
CascadeFormer-F1Params (B)=1.19, Cost (TFLOPs)=4.432026.0617.8851.1686.6523,635
CascadeFormer-C1Params (B)=1.16, Cost (TFLOPs)=4.332026.0617.945381.7125,066
Vanilla-14LParams (B)=1.15, Cost (TFLOPs)=4.272026.061854.6878.0926,227
CascadeFormer-F2Params (B)=1.09, Cost (TFLOPs)=4.052026.0618.149.9281.0625,266
Vanilla-13LParams (B)=1.08, Cost (TFLOPs)=4.002026.0618.1755.1872.428,287
CascadeFormer-C2Params (B)=1.03, Cost (TFLOPs)=3.812026.0618.353.8870.6528,990
Vanilla-12LParams (B)=1.01, Cost (TFLOPs)=3.722026.0618.8255.7466.7530,683