Comment by gscott

It seems FPGA can now do things for LLM's so there might be some future in that

https://www.achronix.com/blog/accelerating-llm-inferencing-f...