vllm serve need chat template.

[pzs19]

System Info / 系統信息

ValueError: 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.

Who can help? / 谁可以帮助到您？

No response

Information / 问题信息

• The official example scripts / 官方的示例脚本
• My own modified scripts / 我自己修改的脚本和任务

Reproduction / 复现过程

vllm serve THUDM/LongWriter-llama3.1-8b
curl http://localhost:8000/v1/completions
-H "Content-Type: application/json"
-d '{
"model": "THUDM/LongWriter-llama3.1-8b",
"prompt": "San Francisco is a",
"max_tokens": 7,
"temperature": 0
}'

Expected behavior / 期待表现

ValueError: 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.