[Runtime] Support PagedKVCache with tree attention #17049
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
MasterJH5574
force-pushed
the
tvm-dev/2024-05-31-paged-kv-tree-attn
branch
from
a51c9c6 to
2ba8a69
Compare
This PR introduces the tree attention to PagedKVCache. With this feature, now the KV cache is ready for tree attention cases such as speculative decoding trees. This PR adds tree attention tests to test the correctness. The changes in this PR to KVState interface are backward compatible.
MasterJH5574
force-pushed
the
tvm-dev/2024-05-31-paged-kv-tree-attn
branch
from
2ba8a69 to
f497859
Compare
tqchen
approved these changes
May 31, 2024
Member
|
@tvm-bot rerun |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR introduces the tree attention to PagedKVCache. With this feature, now the KV cache is ready for tree attention cases such as speculative decoding trees.
This PR adds tree attention tests to test the correctness.
The changes in this PR to KVState interface are backward compatible.