[spec decode] Consolidate speculative decode method name for MTP (#25232)

Signed-off-by: zixi-qi <qizixi@meta.com>
2025-09-26 15:27:05 -07:00
parent cf89202855
commit c70ac4b8ff
6 changed files with 287 additions and 40 deletions
--- a/vllm/engine/arg_utils.py
+++ b/vllm/engine/arg_utils.py
@@ -1481,7 +1481,7 @@ class EngineArgs:
                raise NotImplementedError(
                    "Draft model speculative decoding is not supported yet. "
                    "Please consider using other speculative decoding methods "
-                    "such as ngram, medusa, eagle, or deepseek_mtp.")
+                    "such as ngram, medusa, eagle, or mtp.")

        V1_BACKENDS = [
            "FLASH_ATTN",