Comments (4)
Bot detected the issue body's language is not English, translate it automatically.
Title: I want to ask a newbie question
Let me talk about my situation: I came from the original project and redeployed this one when I saw the plug-in. However, there is no very clear tutorial on how to use it. I have followed the documentation and applied for environment variables in some places. Then I can only use ordinary dialogue and cannot trigger the plug-in.
Then I tested and figured out that it couldn’t be used for a long time, so I analyzed the reasons:
At first I thought this was to first analyze the semantics through gpt and then automatically execute the plug-in based on the keywords. After all, it is langchain and it should have analytical capabilities. I couldn’t follow the documentation and ask questions, so I thought: is this using the official one? The interface is connected to a plug-in that can only be used with an official plus account. Does it require an API4 interface? And I think that only 4 people have permission to use Dalle's official website. This project can be used by filling in CF and does not even need API. It should not be because Dalle is a free public interface, but to apply for docking through key. CF is just a transfer Save resource reverse link. This further confirms this idea.
I don’t know if this is the case, and I don’t have a foreign credit card. I use the built conversation-to-API simulation API. Normal chat can be the same as 3.5. If it is implemented based on the returned semantics, then there should be no problem. But I was surprised when I saw the document saying that except for two models, all other models support plug-ins. So even the 3.5 API can also use this plug-in. Doesn’t the official website not support it? That means that either the API has additional interfaces, or my simulated API actually has incomplete interfaces. But I can't get the key to test. I can only guess at the argument, and it conflicts with the above idea. If 3.5key can use dalle, then why does it need plus membership to use it? It's an endless cycle again.
I see that the issue results seem to be issues raised by experienced users. I’m sorry to bother you by opening an issue, because I’m afraid it’s too low-level. Everyone knows how to use it, so I don’t know if I’ve missed it somewhere. Then I spent another two days looking over the project. Do it again. I even applied for several apikey places, but the SD drawing person couldn't understand it. After looking at it, I feel that it has nothing to do with the official API. Searching requires a search engine key, and drawing requires building an SD. They do not rely on OpenAI. Coming around again it feels like automatically using plugins based on semantics.
I really don't know what's going on, so I thought of posting an issue. If it's really pointless, just close it. It’s not like I just ask questions without taking the time to solve them myself. I’m really a little confused because I don’t play in this industry. I would be very grateful if anyone could give me some pointers on how to use it. If you think my problem is difficult or takes a lot of time, just forget it and tell me and I will go back to the original project.
I'm sorry if I disturbed you all.
from chatgpt-next-web-langchain.
项目使用的是 langchain agent 来实现的插件调用,可以参考这篇文档了解 openai_functions_agent ,核心还是用到了 openai 官方接口的 function call 功能来实现的,这就导致需要你使用官方的 api 接口或者有一个支持 function call
的微调模型接口才能使用本项目的函数调用功能。
from chatgpt-next-web-langchain.
感谢解答 终于解决了我的疑惑
from chatgpt-next-web-langchain.
Bot detected the issue body's language is not English, translate it automatically.
Thanks for the answer. It finally solved my doubts.
from chatgpt-next-web-langchain.
Related Issues (20)
- 同时支持RAG和上传图片 HOT 3
- [Feature Request]: Support image upload for Azure ChatGPT model gpt-4o HOT 2
- [Bug] 调用google搜索异常 HOT 15
- [Bug] google custom search 结果好差 HOT 3
- [Bug] rag功能不是很好使,请指点下 HOT 11
- [Bug] 出现错误,无法回答问题 HOT 11
- [Feature Request]: 先谢!6月7日部署了docker版,与作者官方比无上传图片或者文件按钮,azure中转的api画图后,无法直接保存图片。是否有其他版本可以替换? HOT 10
- azure官方api无法使用 HOT 29
- [Feature Request]: 希望为自定义模型增加插件功能支持 HOT 9
- 求助关于文件上传和社区交流 HOT 8
- 无法使用Google Custom Search HOT 3
- ChatGPT
- [Feature Request]: 自定义支持模型调用的模型列表 HOT 3
- [Feature Request]: 建议增加用户管理能力 HOT 3
- [Feature Request]: Claude3可以接入插件功能吗 HOT 8
- [Feature Request]: 能不能增加自定义模型使用上传文件功能 HOT 11
- [Feature Request]: 增加默认模型的环境变量 HOT 9
- [Bug] 语音转文本提示错误 HOT 26
- [Bug] RAG功能分析文件后,无法下载分析后的文件 HOT 5
- [Bug] Vercel使用RAG的两个小问题 HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from chatgpt-next-web-langchain.