feat(lazy_vlm): implement normal prefill and decode logic#424
Merged
chenghuaWang merged 2 commits intoUbiquitousLearning:v2from Sep 4, 2025
Merged
feat(lazy_vlm): implement normal prefill and decode logic#424chenghuaWang merged 2 commits intoUbiquitousLearning:v2from
chenghuaWang merged 2 commits intoUbiquitousLearning:v2from
Conversation
- Update submodule versions: - fmt: removed branch specification - googletest: removed branch specification - benchmark: removed branch specification - pybind11: removed branch specification - kleidiai: removed branch specification - cccl: removed branch specification and updated to v3.0.2 - cutlass: removed branch specification and updated to v4.1.0 - Fix lazy VLM cache issue by updating hidden state cache correctly - Improve makeWindowIndex function formatting - Add TODO markers for lazy VLM decoding
- Add forward_normal_prefill and forward_normal_decode functions to Qwen2_5VLAttention, Qwen2_5VLDecoder, and Qwen2_5VLText modules - Implement causal mask creation and initialization - Update hidden states and position embeddings for normal prefill and decode - Modify KV cache update logic - Refactor main function to use new normal prefill and decode logic - Update run.py to enable debug build
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
feat(lazy_vlm): implement normal prefill and decode logic