# PACE: Periodic and Channel-Aligned Encoding of Four-Bit Recurrent States > Source: > Published: 2026-10-22 06:30:32+00:00 Linear-attention language models keep their context in a recurrent state of fixed size, where an inference system stores one copy of this state for every sequence it decodes at the same time. Storing the state at lower precision therefore lets more sequences share an accelerator. Existing state quan