QwenLM / Qwen2.5-Coder

Qwen2.5-Coder is the code version of Qwen2.5, the large language model series developed by Qwen team, Alibaba Cloud.

2.94k stars 192 forks source link

the chat template `qwen2_5` corresponding to the model `qwen2_5-coder-32b-instruct-awq` is in chat format. Please use the `chat.completions` API. #178

Open gitYuZui opened 20 hours ago

gitYuZui commented 20 hours ago

部署方式

ms-swift + vllm 方式部署

错误说明：

我从/v1/chat/completions的方式调用进来是正常的。但是从 /v1/completions方式调用，接口会报错：the chat template qwen2_5 corresponding to the model qwen2_5-coder-32b-instruct-awq is in chat format. Please use the chat.completions API.

问题：

1、这是因为这个模型不支持completions方式调用吗？ 2、如何解决？ 3、有没有更好的方式去适配 continue插件？