Hardware
Long-context sparse attention has a catch: data-dependent block selection wrecks memory access kills speed. Our @MiniMax_AI M3 kernel on Bla…
Long-context sparse attention has a catch: data-dependent block selection wrecks memory access kills speed. Our @MiniMax_AI M3 kernel on Blackwell answers it. KV-stationary, each block read once, ~980
Long-context sparse attention has a catch: data-dependent block selection wrecks memory access kills speed. Our @MiniMax_AI M3 kernel on Blackwell answers it. KV-stationary, each block read once, ~980 TFLOP/s on a B200. See the breakdown here → https://fireworks.ai/blog/kernel-optimization-for-minimax-m3-on-nvidia-blackwell
Source: Fireworks AI (X) | 2026-07-10