specific method · filed under model architecture
MQA indexer
A multi-query indexer used to select blocks for sparse attention.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- core
How sources treat it
One count per evidence span, weakest treatment to strongest.
core 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
The selection machinery is small and explicit: an MQA indexer with 4 query heads and a single shared key head, head dimension 128, choosing under a fixed budget of 512 blocks or 2048 tokens.
coremodel architecturein Qwen3.8-Flash-NextQwen
Filed alongside
Other methods under model architecture :: token mixer :: sparse attention :: sparse attention indexer.
Compressed Sparse Attention 2Hierarchical Sparse IndexerIndexCacheIndexShareLightning IndexerReindex ModeReuse ModeDense Warm-up StageFine-grained token selectionIndex BranchIndexer WarmupReuse QSA index selection across speculative decoding stepsAverage poolingBlock-causal scoringCompressed lightweight indexerCross-stage shared-state management for attention reuseDetached indexer-input optimizationIndex Branch outputIndex Branch value headIndexPoolSingle-head index key