gpt4-pdf-chatbot-langchain icon indicating copy to clipboard operation
gpt4-pdf-chatbot-langchain copied to clipboard

Unable to stream response

Open ollieanwyll opened this issue 1 year ago • 2 comments

In the tutorial and demo videos, it is shown that responses are typed back in real-time. However, this functionality seems to be absent in the current main branch, as responses are only displayed in full.

I kindly request assistance from the community in enabling the streaming feature for real-time responses. This would greatly enhance the user experience and align the main branch with the functionality demonstrated in the videos. Any help or guidance would be greatly appreciated. Thank you!

ollieanwyll avatar Apr 28 '23 15:04 ollieanwyll

I noticed it was removed recently myself. I left it in on my fork and love that aspect too. Not sure why it was removed???

https://github.com/groovybits/gaib/blob/main/utils/makechain.ts You can see how to handle streaming in my fork here.

I have the features of the current + streaming and a lot of other additions. I think the streaming aspect is cool and useful. I would love it to come back myself.

What reasons was it removed for?

Also to note that I have it handling multiple name spaces + it retries the GPT API and shows your tokens as they progress with the streaming aspect of it: https://github.com/groovybits/gaib/blob/main/pages/api/chat.ts Which seems really nice.

groovybits avatar Apr 28 '23 15:04 groovybits

Awesome thank you. I'll check it out. Please let me know if you add any more features:)

ollieanwyll avatar Apr 29 '23 06:04 ollieanwyll

Hi, @ollieanwyll! I'm Dosu, and I'm helping the gpt4-pdf-chatbot-langchain team manage their backlog. I wanted to let you know that we are marking this issue as stale.

From what I understand, the issue you raised was about the absence of real-time response streaming in the current main branch. You requested assistance from the community to enable this feature and align the main branch with the functionality demonstrated in the videos. User groovybits noticed the removal of the feature and shared their fork that still has the streaming functionality. They also mentioned additional features they added, such as handling multiple name spaces and showing token progress. You expressed gratitude and interest in any future features added by groovybits.

I wanted to check with you if this issue is still relevant to the latest version of the gpt4-pdf-chatbot-langchain repository. If it is, please let the gpt4-pdf-chatbot-langchain team know by commenting on the issue. Otherwise, feel free to close the issue yourself, or the issue will be automatically closed in 7 days.

Thank you for your contribution, and please don't hesitate to reach out if you have any further questions or concerns.

dosubot[bot] avatar Sep 23 '23 16:09 dosubot[bot]