TensorRT-LLM11 min read
ReplaySSM: Rethinking State Storage for Faster Hybrid-Model Decoding
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...
By Michelle HortonNVIDIA Developer Blog

Read the full story
Originally published by NVIDIA Developer Blog on Oct 12, 2026.
Comments
Sign in to join the discussion.