AMDMIGraphX
AMDMIGraphX copied to clipboard
Use dynamic shapes for kv-cache
There is two parts for this:
- Use dynamic shapes to handle the past sequence length.
- Similar to dynamic batching, we can make two different submodule for each GQA operator for max sequence length and sequence length of 1. Then use a select operator to select the different module based on the inputs sizes(perhaps its possible to reuse the
selectfrom dynamic batching).