diff --git a/vllm_ascend/attention/sfa_v1.py b/vllm_ascend/attention/sfa_v1.py index 1bfe8e67..c4a2a51e 100644 --- a/vllm_ascend/attention/sfa_v1.py +++ b/vllm_ascend/attention/sfa_v1.py @@ -167,7 +167,7 @@ class AscendSFAMetadataBuilder(MLACommonMetadataBuilder[AscendSFAMetadata]): ) -> AttentionCGSupport: # Explicit override in case the underlying builder specialized this getter. # @override omitted only because of mypy limitation due to type variable. - return AttentionCGSupport.UNIFORM_SINGLE_TOKEN_DECODE + return AttentionCGSupport.UNIFORM_BATCH def reorder_batch(self, input_batch: "NPUInputBatch", scheduler_output: "SchedulerOutput") -> bool: