From 767b8dcab31703d7065bd772a7358d8e994ae6b9 Mon Sep 17 00:00:00 2001 From: Gal Hubara-Agam <96368689+galagam@users.noreply.github.com> Date: Wed, 4 Feb 2026 19:46:15 +0200 Subject: [PATCH] [None][chore] AutoDeploy: Set nanov3 and superv3 configs to use flashinfer ssm (#11183) Signed-off-by: Gal Hubara Agam <96368689+galagam@users.noreply.github.com> --- examples/auto_deploy/nano_v3.yaml | 2 ++ examples/auto_deploy/super_v3.yaml | 2 ++ 2 files changed, 4 insertions(+) diff --git a/examples/auto_deploy/nano_v3.yaml b/examples/auto_deploy/nano_v3.yaml index 8844bc228e..4f91b403e0 100644 --- a/examples/auto_deploy/nano_v3.yaml +++ b/examples/auto_deploy/nano_v3.yaml @@ -45,3 +45,5 @@ transforms: fuse_mamba_a_log: stage: post_load_fusion enabled: true + insert_cached_ssm_attention: + backend: flashinfer_ssm diff --git a/examples/auto_deploy/super_v3.yaml b/examples/auto_deploy/super_v3.yaml index 56e7f292da..13b536a630 100644 --- a/examples/auto_deploy/super_v3.yaml +++ b/examples/auto_deploy/super_v3.yaml @@ -44,3 +44,5 @@ transforms: fuse_mamba_a_log: stage: post_load_fusion enabled: true + insert_cached_ssm_attention: + backend: flashinfer_ssm