本地vllm启动的模型，长度超限 || The model started by local vllm has a length exceeding the limit.

400 You passed 74655 input characters and requested 32000 output tokens. However, the model's context length is only 32000 tokens, resulting in a maximum input length of 0 tokens (at most 0 characters). Please reduce the length of the input prompt. (parameter=input_text, value=74655)

1. 本地vllm启动的模型。配置了32K长度，openAI->自定义服务->url方式配置
2. windows10 docker部署。
3. 就问了一句你好 就提示超长了。

---
400 You passed 74655 input characters and requested 32000 output tokens. However, the model's context length is only 32000 tokens, resulting in a maximum input length of 0 tokens (at most 0 characters). Please reduce the length of the input prompt. (parameter=input_text, value=74655)

1. Local vllm startup model. 32K length is configured, openAI->custom service->url configuration
2. windows10 docker deployment.
3. I just asked hello and the prompt was too long.