Skip to content

Commit 63c1ce8

Browse files
committed
Update on "[Executorch][llm] Enable leveraging ring kv cache via module swap"
This allows us to make some of the attention modules to use sliding window kv cache. Will help enable models like gemma3. Differential Revision: [D73891426](https://our.internmc.facebook.com/intern/diff/D73891426/) [ghstack-poisoned]
2 parents 9d46798 + 7638a4b commit 63c1ce8

File tree

0 file changed

+0
-0
lines changed

    0 file changed

    +0
    -0
    lines changed

    0 commit comments

    Comments
     (0)