[Bug] The environment variable DEFAULT_MODEL configuration is invalid

[chiachan163]

📦 Deployment Method

Vercel

📌 Version

v2.15.8

💻 Operating System

Other

📌 System Version

vercel

🌐 Browser

Chrome

📌 Browser Version

131.0.6778.140

🐛 Bug Description

fetch("/api/config", {
        method: "post",
        body: null,
        headers: {
          ...getHeaders(),
        },
      })
        .then((res) => res.json())
        .then((res) => {
          const defaultModel = res.defaultModel ?? "";
          if (defaultModel !== "") {
            const [model, providerName] = getModelProvider(defaultModel);
            DEFAULT_CONFIG.modelConfig.model = model;
            DEFAULT_CONFIG.modelConfig.providerName = providerName as any;
          }

          return res;
        })

配置了环境变量DEFAULT_MODEL=gpt-3.5-turbo@OpenAi ，发现不生效。
看来源码，是有处理的。

最后定位到这里，发现是接口返回数据并没有写入。所以取数据还是默认值gpt-4o-mini

📷 Recurrence Steps

No response

🚦 Expected Behavior

No response

📝 Additional Information

No response

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically.

---

Title: [Bug] The environment variable DEFAULT_MODEL configuration is invalid

[wangping886]

同样的问题，DEFAULT_MODEL 和CUSTOM_MODELS 配置不起作用。我自己本地设置了DEFAULT_MODEL = qwen-turbo
访问web页面看到默认的模型还是gpt-4-mini。 另外我尝试了设置CUSTOM_MODELS = +qwen-7b-chat，也并没有看到新模型；

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically.

---

Same problem, DEFAULT_MODEL and CUSTOM_MODELS configurations don't work. I set DEFAULT_MODEL = qwen-turbo locally myself
Visit the web page and see that the default model is still gpt-4-mini. In addition, I tried setting CUSTOM_MODELS = +qwen-7b-chat, but I did not see the new model;

[wangping886]

同样的问题，DEFAULT_MODEL 和CUSTOM_MODELS 配置不起作用。我自己本地设置了DEFAULT_MODEL = qwen-turbo 访问web页面看到默认的模型还是gpt-4-mini。 另外我尝试了设置CUSTOM_MODELS = +qwen-7b-chat，也并没有看到新模型；

我的是因为配置文件后面的空值DEFAULT_MODEL覆盖了前面的DEFAULT_MODEL 。 而且DEFAULT_MODEL 需要这样设置+qwen-coder-turbo@Alibaba

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically.

---

Same problem, DEFAULT_MODEL and CUSTOM_MODELS configurations don't work. I set DEFAULT_MODEL = qwen-turbo locally and visited the web page to see that the default model is still gpt-4-mini. In addition, I tried setting CUSTOM_MODELS = +qwen-7b-chat, but I did not see the new model;

Mine is because the empty value DEFAULT_MODEL at the end of the configuration file overwrites the previous DEFAULT_MODEL. And DEFAULT_MODEL needs to be set like this +qwen-coder-turbo@Alibaba