Unprecedented
Performance
and Scalability
Performance of HyperAccel LPU
OPT 66B
OPT 30B
OPT 6.7B
OPT 1.3B
23.6
46.5
175.8
520.9
tokens/sec
8x LPU
8:2048
Efficiency Analysis of
HyperAccel LPU vs GPU Platform
Edge
343.8

1x HyperAccel LPU

243.5

1x NVIDIA L4

1.42x
tokens/sec/kW
OPT 6.7B
8:2048
Datacenter
38.8

8x HyperAccel LPU

29.5

2x NVIDIA H100*

1.33x
tokens/sec/kW
OPT 66B
8:2048
* 8x LPU and 2x H100 have similar cost
Scalability Analysis of
HyperAccel LPU vs. GPU Platform
1.8X
1.74X
1.72X
1.8X
1.72X
1.74X