[Bug] 助手名称、描述或者话题名称等“智能生成命名”的能力似乎仅使用openai的模型

[jameszhanghkmci]

💻 系统环境

Windows

📦 部署环境

Vercel / Zeabur / Sealos

🌐 浏览器

Chrome

🐛 问题描述

助手名称、描述或者话题名称等“智能生成命名”的能力似乎仅使用openai的模型，我配置了gemini pro模型但是无法正常生成“智能重命名”有关内容

🚦 期望结果

使用当前助手配置中的模型来进行生成工作

📷 复现步骤

配置一个错误的openai api key，同时配置一个正确的Google Gemini Pro API KEY，且配置当前助手使用gemini pro模型，可以发现智能生成助手信息或者话题信息的能力无法工作

📝 补充信息

No response

[lobehubbot]

👀 @jameszhanghkmci

Thank you for raising an issue. We will investigate into the matter and get back to you as soon as possible.
Please make sure you have given us as much context as possible.
非常感谢您提交 issue。我们会尽快调查此事，并尽快回复您。 请确保您已经提供了尽可能多的背景信息。

[lobehubbot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

💻 System environment

Windows

📦 Deployment environment

Vercel / Zeabur / Sealos

🌐 Browser

Chrome

🐛 Problem description

The ability to "intelligently generate names" such as assistant names, descriptions or topic names seems to only use openai models. I configured the gemini pro model but cannot generate "intelligent renaming" related content normally.

🚦 Expected results

Use the model in the current assistant configuration for the build job

📷 Steps to reproduce

Configure a wrong openai api key, configure a correct Google Gemini Pro API KEY, and configure the current assistant to use the gemini pro model. You can find that the ability to intelligently generate assistant information or topic information cannot work.

📝 Supplementary information

No response

[arvinxx]

目前是还不支持自定义『智能』开头的系列自动化生成功能。后续会提供相应的配置选项

[lobehubbot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

Currently, the automatic generation function of customized series beginning with "Intelligent" is not supported. Corresponding configuration options will be provided later.

[BrandonStudio]

希望同时可以添加一个开关允许禁用智能重命名

[lobehubbot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

Hopefully a switch can also be added to allow disabling smart renaming

[MEIZU16]

我配置了openai的模型（未使用Azure），对话中可以正常使用，但是智能重命名却提示：
**Proxy error (HTTP 402 No Keys Available)**The proxy encountered an error while trying to send your prompt to the upstream service. Further technical details are provided below.***No keys available for model family 'azure-turbo'.{ "error": { "message": "No keys available for model family 'azure-turbo'.", "type": "proxy_no_keys_available" }}

[davidreis97]

What if we just used the current model to generate the title? LMK what you think @arvinxx #1889

[hitzhangjie]

Hope this feature could be supported

[chuangbo]

I also encountered this issue. I hoped to only one local Ollama model to seamless use Lobechat, but unfortunately, the 'Smart Renaming' feature currently requires the use of OpenAI, which presents some inconvenience.

[delicacyyy]

There appears to be an issue with the translation function. Regardless of the model settings I choose, it seems to consistently default to the "gpt-3.5-turbo" model.

[zyfyy]

error message display as the auto renamed title.

[arvinxx]

@zyfyy upgrade to latest version