Add SDPA attention fallback and design docs#231
Open
Lebhoryi wants to merge 3 commits into
Open
Conversation
Qwen3Config may not expose dtype, so fall back to torch_dtype or the current default dtype before initializing model weights and sizing KV cache.
Allow running without flash-attn and document backend selection via NANOVLLM_ATTENTION_BACKEND.
Document the engine architecture, scheduling flow, KV cache lifecycle, prefix cache behavior, attention context, and execution pipeline.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
dtype, falling back totorch_dtypeor the current PyTorch default dtype.flash-attnis unavailable, controlled byNANOVLLM_ATTENTION_BACKEND.Test plan
python -m py_compile nanovllm/engine/model_runner.py nanovllm/layers/attention.py