From a989fef5de1a7f11c219e137944636d1abde1ba4 Mon Sep 17 00:00:00 2001 From: zouyida2052 Date: Wed, 22 Oct 2025 21:55:06 +0800 Subject: [PATCH] unify logic between aclgraph and torchair (#3602) ### What this PR does / why we need it? unify logic between aclgraph and torchair. This is a cherry-pick of https://github.com/vllm-project/vllm-ascend/pull/3560 Signed-off-by: zouyida2052 --- vllm_ascend/spec_decode/mtp_proposer.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm_ascend/spec_decode/mtp_proposer.py b/vllm_ascend/spec_decode/mtp_proposer.py index a950954..2a11731 100644 --- a/vllm_ascend/spec_decode/mtp_proposer.py +++ b/vllm_ascend/spec_decode/mtp_proposer.py @@ -502,7 +502,7 @@ class MtpProposer(Proposer): # prepare next mtp inputs # mtp>1: prefill skip or decode skip last loop - if with_prefill and self.torchair_graph_enabled: + if with_prefill: for _ in range(self.num_speculative_tokens - 1): draft_token_ids_list.append(draft_token_ids) if step == self.num_speculative_tokens - 1 or with_prefill: