vllm - 💡(How to fix) Fix [Bug]: vllm.entrypoints.chat_utils.ChatTemplateResolutionError: As of transformers v4.44, default chat template is no longer allowed, so you must provide a chat template if the tokenizer does not define one. [1 participants]

Your current environment

vllm:v0.19.0

🐛 Describe the bug

using vllm 0.19.0，upgrade transformers to 5.5.0
- - command: - /bin/sh - -c - | pip install transformers>=5.5.0
exec python3 -m vllm.entrypoints.openai.api_server
--host "0.0.0.0"
--port "8000"
--model /models/gemma-4-31B/
--served-model-name local-gemma4-31b
--enable-chunked-prefill
--gpu-memory-utilization "0.87"
--tensor-parallel-size "4"
--trust-remote-code
--max-model-len "131072"
--uvicorn-log-level warning
--enable-auto-tool-choice
--tool-call-parser gemma4
--reasoning-parser gemma4
--limit-mm-per-prompt '{"image": 15, "audio": 1}'
--async-scheduling
--default-chat-template-kwargs '{"enable_thinking": true}'

request error: vllm.entrypoints.chat_utils.ChatTemplateResolutionError: As of transformers v4.44, default chat template is no longer allowed, so you must provide a chat template if the tokenizer does not define one.

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.