Comment Comment by JohannaAlmeida Full attention O(n²): 17.96s / 5.6 tok/sHybridAttention O(n·W + n·D): 0.35s / 286.6 tok/s parent JohannaAlmeida 7 months ago
HybridAttention O(n·W + n·D): 0.35s / 286.6 tok/s