diff --git a/server/text_generation_server/models/flash_causal_lm.py b/server/text_generation_server/models/flash_causal_lm.py index c44dd57d7..f332ab51d 100644 --- a/server/text_generation_server/models/flash_causal_lm.py +++ b/server/text_generation_server/models/flash_causal_lm.py @@ -225,6 +225,7 @@ class FlashCausalLMBatch(Batch): return FlashCausalLMBatch( batch_id=self.batch_id, + past_pad=self.past_pad, requests=requests, requests_idx_mapping=requests_idx_mapping, input_ids=input_ids, @@ -311,6 +312,7 @@ class FlashCausalLMBatch(Batch): return FlashCausalLMBatch( batch_id=batches[0].batch_id, + past_pad=batches[0].past_pad, requests=requests, requests_idx_mapping=requests_idx_mapping, input_ids=input_ids,