Support topk > 1 tree drafting for mamba/hybrid-linear models on spec v2 (#27463)

2026-06-07 17:04:09 -07:00
parent 70db73afce
commit f68c79675f
8 changed files with 45 additions and 57 deletions
@@ -14,6 +14,9 @@ QWEN3_NEXT_MODEL = "Qwen/Qwen3-Next-80B-A3B-Instruct"
 class TestQwen3NextMTPTopk(
    GSM8KMixin, KLDivergenceMixin, PrefixCacheBranchingMixin, DefaultServerBase
 ):
+    # topk > 1 (tree) MTP on a hybrid-GDN model, on spec v2: the tree-aware mamba
+    # state update lives in the spec v2 verify path, so mamba + topk > 1 no longer
+    # falls back to spec v1.
    model = QWEN3_NEXT_MODEL
    cache_chunk_size = 64
    gsm8k_accuracy_thres = 0.93