Copilot + GPT-5.2 超神组合!接入第三方API

先说优点,比cc好用!比 codex 快!代码质量也比cc好得多!

配置说明:

安装 copilot

安装 copilot 接入第三方API的插件


{
  "reasoning": {
    "effort": "xhigh",
    "summary": "auto"
  },
  "store": false,
  "stream": true
}

effort 可选项 :lowmediumhighxhigh

配置密钥:

这里前面的小眼睛要点一下显示!

然后切换一下模型就可以开始愉快的玩帅了!

最后贴一个 settings.json 的配置版本

我这里的baseUrl 用的是 cc-switch 的proxy 地址,密钥随便填就行!

"gcmp.compatibleModels": [
    {
        "id": "rc:gpt-5.2",
        "name": "RC-GPT-5.2-Xhigh",
        "provider": "zhipu",
        "baseUrl": "http://127.0.0.1:15721",
        "model": "gpt-5.2",
        "sdkMode": "openai-responses",
        "maxInputTokens": 272000,
        "maxOutputTokens": 128000,
        "capabilities": {
            "toolCalling": true,
            "imageInput": true
        },
        "useInstructions": false,
        "includeThinking": true,
        "extraBody": {
            "reasoning": {
                "effort": "xhigh",
                "summary": "auto"
            },
            "store": false,
            "stream": true
        }
    },
    {
        "id": "rc:gpt-5.2-low",
        "name": "RC-GPT-5.2-Low",
        "provider": "zhipu",
        "baseUrl": "http://127.0.0.1:15721",
        "model": "gpt-5.2",
        "sdkMode": "openai-responses",
        "maxInputTokens": 272000,
        "maxOutputTokens": 128000,
        "capabilities": {
            "toolCalling": true,
            "imageInput": true
        },
        "useInstructions": false,
        "includeThinking": true,
        "extraBody": {
            "reasoning": {
                "effort": "low",
                "summary": "auto"
            },
            "store": false,
            "stream": true
        }
    },
     {
        "id": "rc:gpt-5.2-High",
        "name": "RC-GPT-5.2-High",
        "provider": "zhipu",
        "baseUrl": "http://127.0.0.1:15721",
        "model": "gpt-5.2",
        "sdkMode": "openai-responses",
        "maxInputTokens": 272000,
        "maxOutputTokens": 128000,
        "capabilities": {
            "toolCalling": true,
            "imageInput": true
        },
        "useInstructions": false,
        "includeThinking": true,
        "extraBody": {
            "reasoning": {
                "effort": "high",
                "summary": "auto"
            },
            "store": false,
            "stream": true
        }
    }
],
63 个赞

这个好 感谢分享

3 个赞

试试看的

2 个赞

感谢佬友,又多了一个方式使用,get新知识

4 个赞

跟cline比较过,还是cline给出的结果符合我需要的,所以还是用cline了。

1 个赞

感谢分享

3 个赞

结合5.2很强,主要看模型

1 个赞

copilot自动压缩吗 没有用过

2 个赞

自动压缩的,不用关心压缩

1 个赞

感谢分享!

2 个赞

copilot自带的gpt5.2不可以吗qw

1 个赞

也可以,但是贵呀!

4 个赞

感谢大佬!

1 个赞

收藏备用

1 个赞

大佬牛逼

1 个赞

不愧是天才程序员

1 个赞

感谢大佬分享

1 个赞

无事了,是我自己的上游问题

2 个赞

coplite可以的,爽用

1 个赞

可以,收了,感谢佬

1 个赞