From 97d857c096bf4ca9ba0be1e8d92b96efb3d75e82 Mon Sep 17 00:00:00 2001 From: Hanming Lu <69857889+hanming-lu@users.noreply.github.com> Date: Wed, 15 Oct 2025 18:56:43 -0700 Subject: [PATCH] [Mamba] Increase default mamba_full_memory_ratio to 0.9 (#11679) --- python/sglang/srt/server_args.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/python/sglang/srt/server_args.py b/python/sglang/srt/server_args.py index dbfd48efb..b5ae57817 100644 --- a/python/sglang/srt/server_args.py +++ b/python/sglang/srt/server_args.py @@ -371,7 +371,7 @@ class ServerArgs: # Mamba cache max_mamba_cache_size: Optional[int] = None mamba_ssm_dtype: str = "float32" - mamba_full_memory_ratio: float = 0.2 + mamba_full_memory_ratio: float = 0.9 # Hierarchical cache enable_hierarchical_cache: bool = False