### Your current environment <details> <summary>The output of `python collect_env.py`</summary> ```text Your output of above commands here ``` </details> ### 🐛 Describe the bug 我根据 https://vllm-ascend.readthedocs.io/en/latest/tutorials/multi_npu_qwen3_next.html 的教程用v0.11.0rc0部署了Qwen3-Next-80B-A3B-Instruct模型,正常对话精度没有问题,但是当问的问题中有数字的时候,则会出现胡言乱语的情况 <img width="3384" height="1222" alt="Image" src="https://github.com/user-attachments/assets/a61fbea9-5c45-49a1-8e21-c96ec60282ea" />