让ollama调用英伟达显卡gpu
如果你的系统是Ubuntu22.04这个方法可以帮助你，如果你安装了 显卡驱动和 CUDA Toolkit就可以使用以下代码替换你Ollama容器应用的compose 文件在参数的高级设置里面。注意把代码复制到txt文本文档，并将符号`替换为空格符号再把注释删除就可以直接使用，否则缩进不正确会报错。
networks:
````1panel-network:
````````external:`true
services:
````ollama:
````````container_name:`${CONTAINER_NAME}
````````deploy:
````````````resources:
````````````````limits:
````````````````````cpus:`${CPUS}
````````````````````memory:`${MEMORY_LIMIT}
````````````````reservations:# 从这里开始到count: 1就是调用gpu的参数，这里不要修改。
````````````````````devices:
````````````````````````-`capabilities:
````````````````````````````-`gpu
``````````````````````````count:`1
````````environment:
````````````OLLAMA_MAX_LOADED_MODELS:`1# 这里是设置多模型并发的，这里我设置的是1个模型可以根据你的需要增加，但增加会影响性能。
````````````OLLAMA_NUM_PARALLEL:`10# 这里是设置单个模型可以同时回答问题的数量，这里我设置的是可以同时回答10个可以根据你的需要增加，但增加会影响性能。
````````image:`ollama/ollama:0.1.38# 这里是Ollama的版本号要改为你当前的ollama版本号。
````````labels:
````````````createdBy:`Apps
````````networks:
````````````-`1panel-network
````````ports:
````````````-`${HOST_IP}:${PANEL_APP_PORT_HTTP}:11434
````````restart:`unless-stopped
````````tty:`true
````````volumes:
````````````-`./data:/root/.ollama