diff --git a/deploy/mavgpud.json b/deploy/mavgpud.json index fb33c2f..41d001d 100644 --- a/deploy/mavgpud.json +++ b/deploy/mavgpud.json @@ -15,6 +15,7 @@ "--batch-size", "2048", "--ubatch-size", "512", "--jinja", + "--chat-template-kwargs", "{\"enable_thinking\":false}", "--spec-type", "draft-mtp", "--spec-draft-n-max", "2" ],