[ssivark]

When aiming for 100k tok/s, you would still have CUDA overheads (on the order of microseconds) -- which might become the bottleneck, even if you do everything else right with the inference architecture. How are you planning to overcome that?

EDIT: Oh, on second read, do you mean you're running the model on an FPGA?

[taneq]

You might be conflating throughput with latency. 100k tok/s is very different to 1 tok/10us.

[ssivark]

When doing auto regressive inference, how often do you do a CUDA kernel call? What is the main bottleneck at the throughputs you're operating?