From 24bcbc5087ba981218cccbdb7532609a037cff75 Mon Sep 17 00:00:00 2001 From: MOVZX Date: Sun, 27 Sep 2026 00:07:04 +0200 Subject: [PATCH] recoverssm: RecoverSSM for Qwen GDN and the PLE short conv (port of upstream #58863, 03cae7837f) Port of the upstream vLLM change #58863 (Apache-2.0). With --use-replayssm and speculative decoding, Gated DeltaNet layers verify the draft window from a read-only checkpoint and keep a per-token replay record; after sampling one commit replays the accepted tokens and writes the recurrent state once. The PLE short conv gets the same treatment. Co-Authored-By: Hinata Signed-off-by: MOVZX --- vllm/config/cache.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm/config/cache.py b/vllm/config/cache.py index b899502165..9ce836ef15 100644 --- a/vllm/config/cache.py +++ b/vllm/config/cache.py @@ -212,7 +212,7 @@ class CacheConfig: mamba_block_size is a multiple of replayssm_buffer_len, but this is not required.""" use_recoverssm: bool = field(default=False, init=False) - """Whether Kimi-K3 KDA uses RecoverSSM speculative decode.""" + """Whether RecoverSSM speculative decode is used (Kimi-K3 KDA; Qwen4Exp GDN and PLE).""" # Will be set after profiling. num_gpu_blocks: int | None = field(default=None, init=False)