DeepSeek-V4
DeepSeek-V4 — CSA and HCA hybrid attention, mHC, Muon, 32T+ tokens
Why it matters
At 27% of the per-token FLOPs and 10% of the KV cache of V3.2, it made 1M context something supported all the time
Source — Primary sources first; secondary reports are labelled “Press”
Paperarxiv.orghttps://arxiv.org/abs/2606.19348Added in Week 35, 2026 · See that week