单机多卡跑gradio推理时，报CUDA的错误

[chelovek21]

问题

单机多卡【4*RTX4090 24G】gradio推理时，加载模型可以成功，但问答时报CUDA的错误，能提供一下您运行的基础环境或可能的解决思路吗？谢谢
RuntimeError: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

基础环境

• [1] Linux python3.10.8
• [2] pytorch=1.13.1
• [3] transformers=4.29.1
• [4] accelerate=0.20.3
• [5] peft=0.3.0