Update compressed_tensors_w8a8_fp8.py

HabanaAI · Sep 25, 2024 · 4cf34f4 · 4cf34f4
1 parent 426e8e1
commit 4cf34f4
Showing 1 changed file with 1 addition and 0 deletions.
diff --git a/...el_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py b/...el_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py
@@ -13,6 +13,7 @@
 from vllm.model_executor.parameter import (ChannelQuantScaleParameter,
                                            ModelWeightParameter,
                                            PerTensorScaleParameter)
+from vllm.platforms import current_platform
 from vllm.utils import is_hip
 
 __all__ = ["CompressedTensorsW8A8Fp8"]