allow requests with exactly context_len total tokens (#22546)

Co-authored-by: Ethan (Yusheng) Su <yushengsu.thu@gmail.com>
2026-04-30 01:12:06 -07:00
parent 125f75db72
commit 99c0b62f1e
1 changed files with 1 additions and 1 deletions
@@ -842,7 +842,7 @@ class TokenizerManager(TokenizerControlMixin, TokenizerManagerScoreMixin):
        if (
            self.validate_total_tokens
            and max_new_tokens is not None
-            and (max_new_tokens + input_token_num) >= _max_req_len
+            and (max_new_tokens + input_token_num) > _max_req_len
        ):
            if self.server_args.allow_auto_truncate:
                logger.warning(