Forum Discussion

lexie11's avatar
lexie11
Icon for New Contributor rankNew Contributor
11 months ago

Agilex7 m-series for llama

I am undertaking a project to deploy llama using the agilex7 m-series, and during the process, I utilized the FPGA AI Suite. However, dla_compiler does not support the sinking of graphs to FPGA. Coul...
raghad's avatar
raghad
Icon for New Contributor rankNew Contributor
3 months ago

Hi lexie11​ ,

I wanted to follow up on your post and ask about the limitations you encountered.

Were you eventually able to get LLM inference running on the Agilex board?

Did you find any alternative approach such as custom RTL, oneAPI kernels, or another toolchain that worked better than FPGA AI Suite for transformer models?

If you are open to it, having an email connection would also be really helpful for a deeper discussion.

Thanks in advance — any insight would be really appreciated.