Tencent Youtu's LightningMoBA speeds long-context attention 1.2-1.6x without quality loss
tencent · hf · 2026-11-20
Tencent Youtu Lab introduces LightningMoBA, co-designing selection and execution for Mixture of Block Attention via dynamic routing (thresholded per-query block scores emitting variable-cardinality sparse metadata), KV-centric ownership (grouping by query head/KV block, splitting high-fanout columns, reusing loaded KV), and runtime pipelining that overlaps K/V transfers with Tensor Core compute on Hopper GPUs. At matched Top-16 retention across five checkpoints, end-to-end generation accelerates 1.20-1.60x while mean RULER rises from 74.42 to 75.28.
More from Infra
- Starlink Mini enables remote water gate control in mountainous Japan; all 6 tests successful — XFreeze · 2026-11-20
- Jensen Huang's personal AI clout won't transfer to Nvidia's next CEO, analyst says — FinanceYF5 · 2026-11-20
- Europe's AI test happens inside the factory: sovereign compute must fit installed operations — ingliguori · 2026-11-20
- Mubit + Neon: agent middleware cuts frontier-model tokens by 73% — JeremyCMorgan · 2026-11-20
- 2DSD Speculative Decoding Speeds Up Looped LMs up to 7.64x With Lower KV Memory — LucaAmb · 2026-11-20
- US House Reveals 3.4M Chip Export Applications to China, Including 445,000 NVIDIA H200s — Dan_Jeffries1 · 2026-11-20