From 07957e12f45611363970f717dd6be080f5c7b71c Mon Sep 17 00:00:00 2001
From: bastefaniak <bstefaniak@nvidia.com>
Date: Fri, 17 Jul 2026 00:01:21 +0200
Subject: [PATCH] Fix FP8 Quack GEMM call when inference_mode() and
 torch.compile() is used. (#5153)

Signed-off-by: Bartosz Stefaniak <bstefaniak@nvidia.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
---
 vllm_omni/quantization/quack_fp8.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/vllm_omni/quantization/quack_fp8.py b/vllm_omni/quantization/quack_fp8.py
index 8c7251e43..865958386 100644
--- a/vllm_omni/quantization/quack_fp8.py
+++ b/vllm_omni/quantization/quack_fp8.py
@@ -94,7 +94,7 @@ def quack_scaled_fp8_mm(
         return None
     out = torch.empty(a.shape[0], b.shape[1], device=a.device, dtype=out_dtype)
     alpha = scale_a.reshape(1).float() * scale_b.reshape(1).float()
-    gemm.gemm_tuned(a, b, out, bias=bias, alpha=alpha)
+    gemm.gemm(a, b, out=out, bias=bias, alpha=alpha, tuned=True)
     return out