From 1c04d6078bd68686270eeb4d9d28865663d3290a Mon Sep 17 00:00:00 2001 From: laoyao0822 Date: Wed, 25 Mar 2026 21:12:30 +0800 Subject: [PATCH] feat(token_limit): limit input token < 131072 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 4.max_tokens=131072，input＜131072时不报错 5. 输入长度超过131072时，报错，且报错内容含有以下字符串，以便被网关识别 "is longer than the model's context length" --- python/sglang/srt/managers/tokenizer_manager.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/python/sglang/srt/managers/tokenizer_manager.py b/python/sglang/srt/managers/tokenizer_manager.py index 4550b0381..5d5d1a1df 100644 --- a/python/sglang/srt/managers/tokenizer_manager.py +++ b/python/sglang/srt/managers/tokenizer_manager.py @@ -797,7 +797,7 @@ class TokenizerManager(TokenizerCommunicatorMixin, TokenizerManagerMultiItemMixi # Validate input length if input_token_num >= self.context_len: - if self.server_args.allow_auto_truncate: + if False: logger.warning( f"The input ({input_token_num} tokens) is longer than the " f"model's context length ({self.context_len} tokens). " @@ -818,7 +818,7 @@ class TokenizerManager(TokenizerCommunicatorMixin, TokenizerManagerMultiItemMixi and max_new_tokens is not None and (max_new_tokens + input_token_num) >= _max_req_len ): - if self.server_args.allow_auto_truncate: + if True: logger.warning( f"Requested token count ({input_token_num} input + {max_new_tokens} new) " f"exceeds the model's context length ({self.context_len} tokens). "