[Bug] 在 OpenAI 模型中添加了 gemini 1.5 带有视觉的模型，在聊天页面不支持粘贴剪贴板中的图片，手动上传图片可以识别

[allanpk716]

📦 部署环境

Docker

📌 软件版本

v1.45.4

💻 系统环境

Windows

🌐 浏览器

Edge

🐛 问题描述

在 OpenAI 模型中添加了 gemini 1.5 带有视觉的模型（确认添加的时候勾选了 vision 选项），在聊天页面不支持粘贴剪贴板中的图片，手动上传图片可以识别。如果是直接使用 OpenAI 4o mini 则可以粘贴剪贴板中的图片。

📷 复现步骤

在 OpenAI 模型中添加了 gemini 1.5 带有视觉的模型（确认添加的时候勾选了 vision 选项），在聊天页面不支持粘贴剪贴板中的图片

🚦 期望结果

在 OpenAI 模型中添加了 gemini 1.5 带有视觉的模型（确认添加的时候勾选了 vision 选项），在聊天页面支持粘贴剪贴板中的图片

📝 补充信息

No response

[lobehubbot]

👀 @allanpk716

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. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

📦 Deployment environment

Docker

📌 Software version

v1.45.4

💻 System environment

Windows

🌐 Browser

Edge

🐛 Problem description

The gemini 1.5 model with vision has been added to the OpenAI model (confirm that the vision option is checked when adding). Pasting images in the clipboard is not supported on the chat page. Manually uploading images can be recognized. If you use OpenAI 4o mini directly, you can paste the image in the clipboard.

📷 Steps to reproduce

The gemini 1.5 model with vision has been added to the OpenAI model (confirm that the vision option is checked when adding). Pasting images in the clipboard is not supported on the chat page.

🚦 Expected results

The gemini 1.5 model with vision has been added to the OpenAI model (confirm that the vision option is checked when adding), and supports pasting of images in the clipboard on the chat page.

📝 Supplementary information

No response

[dosubot]

Hey @allanpk716, I'm here to help you with any bugs, questions, or contributions you might have. Feel free to ask me anything! 🤖

The issue you're experiencing with the Gemini 1.5 model not supporting clipboard image pasting, while manual uploads work, might be related to a known limitation. The Gemini 1.5 model has a documented issue where it cannot input images when the server database is enabled ^[1]. This might be affecting the clipboard paste functionality as well.

If the OpenAI 4o mini model supports clipboard pasting, it suggests that the issue is specific to the Gemini 1.5 model's configuration or limitations. You might want to check if there are any specific conditions or configurations in the code that disable clipboard image pasting for the Gemini 1.5 model, although I couldn't find specific details on this in the repository.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[GowayLee]

同样有发生在 #5358 中, 不知道有没有成功修复.

[lobehubbot]

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

---

The same is happening in #5358, I don’t know if it has been successfully fixed.