diff --git a/src/llmcompressor/pipelines/sequential/helpers.py b/src/llmcompressor/pipelines/sequential/helpers.py
index e13144689..273ed76d4 100644
--- a/src/llmcompressor/pipelines/sequential/helpers.py
+++ b/src/llmcompressor/pipelines/sequential/helpers.py
@@ -538,6 +538,8 @@ def dispatch_for_sequential(model: PreTrainedModel) -> PreTrainedModel:
 
     if torch.cuda.is_available():
         offloaded_dispatch(model, execution_device=torch.device("cuda:0"))
+    elif hasattr(torch, "npu") and torch.npu.is_available():
+        offloaded_dispatch(model, execution_device=torch.device("npu:0"))
     elif hasattr(torch, "xpu") and torch.xpu.is_available():
         offloaded_dispatch(model, execution_device=torch.device("xpu:0"))
     else: