[Feature] 关于“面具”能力全面向“GPTs”看齐的建议 #3217
Comments
Title: [Feature] Suggestions on “Mask” capabilities being fully aligned with “GPTs”
**Is this feature related to an existing issue? ** **What features do you want or have any suggestions? ** Setting preset words for each conversation, customizing model parameters, Zero One and Few Shot, etc. are functions that have been implemented for a long time. Now we only need DALL-E 3, GPT-4 V and Assistants API. Today I had a whim and thought of a few more suggestions: First, set up two separate top tabs for DALL E and Assistants, which only provide the ability to generate pictures and upload/analyze files, 🔗[because](https://platform.openai.com/docs/assistants/how-it-works/ limitations) In the next few months, the Assistants API will integrate DALL E and GPT-4 V, so multi-modal capabilities can be completed within the conversation only through the Assistants API. After the integration is completed, add Assistants to the global or conversation settings to become a permanent guest. Use the switch button to turn on/off whether to enable Vincentian pictures, whether to allow uploading of pictures, whether Functions such as uploading files are allowed, and corresponding function parameters can be set separately (similar to the preset prompt words of GPTs...). After the button is turned on, the corresponding UI controls will be displayed in the dialog, which also provides users with With a huge choice space, mask will have full multi-modal capabilities by then, and yes, there will also be voice functions such as TTS. Regarding Internet access capabilities, here is a 🔗reference **Are there any similar competing products that we can refer to? ** other information |
|
很好的想法，这样面具就类似 assistant，而插件类似于官网的插件，但是这样似乎和官方的能力拉不开差距？（还是官方太猛了） |
It's a very good idea, so that the mask is similar to the assistant, and the plug-in is similar to the plug-in on the official website, but this seems to be unable to close the gap with the official capabilities? (Official is still too fierce) |
对，该接入的接入，毕竟人家提供 API 了，拉开差距可能要 3.0 版的插件😋 |
Yes, the access should be accessed. After all, they provide API. To widen the gap, you may need version 3.0(#138 (comment)) Plugin 😋 |
我觉得不需要拉开差距，把这些弄完就很丰富了，咱们用 API 的魔力在于随时想用就用，想玩就玩，每月又不用花太多钱，需求大的早都开 Plus 去了，毕竟无限用，使用门槛又低。 |
I don’t think there is a need to widen the gap. Once these are completed, it will be very rich. The magic of us using API is that we can use it whenever we want and play whenever we want. We don’t have to spend too much money every month. If there is a big demand, we can open Plus early. Go, after all, it has unlimited uses and the threshold for use is low. |
|
also for code interpreting since we already had mask, it only need file upload system Desktop.2023.11.11.-.23.08.45.04.mp4 |
|
another show case about code interpreting LLM Api (this a wilds)
|
|
Assistant API 使用的7个例子，可以参考这里： |
For 7 examples of Assistant API usage, please refer here: |
|
要不先把文件上传功能放上去，嘿嘿，API 还是快很多，比官网体验好多了 |
Why not put the file upload function first? Hehe, the API is still much faster and the experience is much better than the official website. |
My side supports the gpt4-vision-preview image recognition function, you can learn more about it. |
This comment was marked as abuse.
This comment was marked as abuse.
I have added gpts store support here. If you need it, you can learn more: https://github.com/vual/ChatGPT-Next-Web-Pro |
This comment was marked as duplicate.
This comment was marked as duplicate.
这个功能与现有的问题有关吗？
、
你想要什么功能或者有什么建议？
看到推上分享的 GPTs 功能，非常惊讶，这不就是“面具”吗，合理怀疑 CloseAI 抄袭本项目创意。
给每个对话设置预设词、定制模型参数、Zero One and Few Shot 等是早已实现的功能，现在就差 DALL-E 3、GPT-4 V 以及 Assistants API，今天突发奇想，又稍微想到几个建议：
先单独设置两个置顶 Tab 给 DALL E 和 Assistants，仅提供生图和上传/分析文件的能力，🔗因为 未来几个月 Assistants API 会将 DALL E 和 GPT-4 V 融合进来，所以到时候仅通过 Assistants API 就可以在对话内完成多模态能力。
等融合完毕之后，再把 Assistants 加到全局或者对话设置里，成为常驻嘉宾，通过开关按钮，来开启/关闭是否启用文生图、是否允许上传图片、是否允许上传文件等功能，同时可以分别设置相应的功能参数（类似 GPTs 的预设提示词啥的...），按钮打开后，相应的 UI 控件才会显示到对话中，这样也给用户极大地选择空间，到时候面具就会全面拥有多模态能力，哦对，还有 TTS 等语音功能。
关于上网能力，这里有一个🔗参考
关于 Assistants API 的能力扩展（接入各种知识库的开放 API），这里有一个🔗列表
有没有可以参考的同类竞品？
、
其他信息
、
The text was updated successfully, but these errors were encountered: