@iplayfast the tools option helps you more convenient or without tools option you can use prompts like { "type": "function", "function": { "name": "get_stock_price", "description": "Get the stock price of an array of stocks", "parameters": { "type": "object", "properties": { "names": { "type": "array", "items": { "type": "string" }, "description": "An array of stocks" } }, "required": [ "names" ] } } }
3条答案
按热度按时间p8h8hvxi1#
Ollama没有工具，只有工具支持。正如示例所示，最终使用LLM的工具和实现取决于您构建这些工具。
yvgpqqbh2#
@iplayfast the tools option helps you more convenient or without tools option you can use prompts like
{
"type": "function",
"function": {
"name": "get_stock_price",
"description": "Get the stock price of an array of stocks",
"parameters": {
"type": "object",
"properties": {
"names": {
"type": "array",
"items": {
"type": "string"
},
"description": "An array of stocks"
}
},
"required": [
"names"
]
}
}
}
dba5bblo3#
工具支持为模型提供一组可能被"调用"的工具功能列表。模型生成一个包含函数请求和参数值的响应。接收到响应的客户端将解析响应并找到工具调用。然后，客户端将使用请求的参数调用该函数。接着，客户端从函数调用中获取响应，并添加一条新消息，角色为：工具，内容为工具的响应。然后将更新后的聊天发送给模型。模型会像收到工具执行结果一样读取这个响应。
客户端需要代表Ollama模型调用这些工具。