From de08b8f61bd412f2bcc3a5f21da26871e39a9078 Mon Sep 17 00:00:00 2001 From: Jinzhen Lin Date: Fri, 19 Dec 2025 12:29:48 +0800 Subject: [PATCH] [Quantization] enable compressed-tensors marlin support for turing (#31000) Signed-off-by: Jinzhen Lin --- .../layers/quantization/kernels/mixed_precision/marlin.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm/model_executor/layers/quantization/kernels/mixed_precision/marlin.py b/vllm/model_executor/layers/quantization/kernels/mixed_precision/marlin.py index faaa45b861de7..eb14f9ec378c4 100644 --- a/vllm/model_executor/layers/quantization/kernels/mixed_precision/marlin.py +++ b/vllm/model_executor/layers/quantization/kernels/mixed_precision/marlin.py @@ -30,7 +30,7 @@ from .MPLinearKernel import MPLinearKernel, MPLinearLayerConfig class MarlinLinearKernel(MPLinearKernel): @classmethod def get_min_capability(cls) -> int: - return 80 + return 75 @classmethod def can_implement(cls, c: MPLinearLayerConfig) -> tuple[bool, str | None]: