NVIDIA Developer Blog discusses ReplaySSM, a state-storage approach for faster decoding in hybrid models that mix state-space model (SSM) layers with a few attention layers. The excerpt names Nemotron 3, Qwen3.5 through Qwen3.8, Kimi-K3, and GLM-5.3-Flash.

Hybrid models interleave state-space model (SSM) layers with a few attention layers. NVIDIA Nemotron 3, Qwen3.5 through Qwen3.8, Kimi-K3, and GLM-5.3-Flash are...