Bot generation code very slow

[f33rni]

Bot generation code very slow

[vuisme]

explorer_NwM9NxsOCm_4.mp4

[Deplyapp]

How did you use the chatgpt it's not working for me can you please tell me how to use it.

[vuisme]

How did you use the chatgpt it's not working for me can you please tell me how to use it.

I'm using other proxy: https://github.com/pandora-next/deploy

[Deplyapp]

How did you use the chatgpt it's not working for me can you please tell me how to use it.

I'm using other proxy: https://github.com/pandora-next/deploy

I have to host this

[vuisme]

My method is: Access Token (take from https://chat.openai.com/api/auth/session)
Proxy url: http://host/backend-api/

[F33RNI]

explorer_NwM9NxsOCm_4.mp4

Why do I think this is slow?
Telegram doesn't allow to fastly spam messages, so I added a small delay that you can change in the config file

[F33RNI]

Thanks a lot for your help. I have tried it but it not work I got the screenshot of my config file please tell me it iam doing something wrong.

My method is: Access Token (take from https://chat.openai.com/api/auth/session) Proxy url: http://host/backend-api/

Please create a new issue for this. Because it's a different topic

[F33RNI]

Bot generation code very slow and almost time out, even though you've set a 480-second timeout. Text generation works fine

Try edit these lines

https://github.com/F33RNI/GPT-Telegramus/blob/master/config.json#L158-L159

[vuisme]

explorer_NwM9NxsOCm_4.mp4

Why do I think this is slow? Telegram doesn't allow to fastly spam messages, so I added a small delay that you can change in the config file

I will try. It's slow because on chatgpt, the answer is complete with full answer but on chatbot it's continues add word per word. It's should be use all text on answer insert to reply, not word per word

[F33RNI]

If you know another method for sending streaming text to Telegram bot without triggering a spam exception, please suggest your solution