Xiaomi Details HySparse2 for MiMo-V3: Two-Level KV Sharing Cuts 1M-Token Prefill FLOPs About 5× vs Hybrid SWA
Xiaomi's LLM-Core team published HySparse2, a hybrid sparse-attention design with two-level KV sharing aimed at MiMo-V3-class agentic models, reporting about 5× lower 1M-token prefill FLOPs versus Hybrid SWA on an 80B-A3B MoE.
Read the full story at Pandaily ↗