Add flash decoding(flash attention with split_kv) #17
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Outdated suggestions cannot be applied.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.
Implement flash decoding(flash attention with split_kv) https://princeton-nlp.github.io/flash-decoding/. This algorithm is used when
batch_size * num_heads * blocks_along_seqlen_q
cannot saturate the gpu's SM's.Benchmark results o RTX-3090.
batch_size=2, num_heads=32, seqlen_q =1, seqlen_k=N_CTX
batch_size=2, num_heads=16, seqlen_q =1, seqlen_k=N_CTX