Create REST API using NestJS(NodeJS) and ChatGPT

[@MagicHappyPotato]

Wow, it's really useful video. Thanks for sharing!

[@gokselozkazanc2917]

Thanks a lot, I followed the steps and it worked.
Could you please provide the test file for the service? I tried it but could not make it work.

[@computingpower]

Hello! Sure, I can make some unit tests and end-to-end tests for the application and share them in the Github repository :)!

[@gokselozkazanc2917]

@@computingpower Thanks a lot!
Could you also share the github repo. I could not find it here in the description.

[@computingpower]

@@gokselozkazanc2917 github.com/Mpcsj-Computing/NestJsRestAPIChatGpt

[@computingpower]

I just updated the main branch of the repository to include unit tests and end-to-end tests. You can run "yarn test" to run unit and integration tests and "yarn test:e2e" to run end-to-end tests

[@wassimbenr]

Great video! just one question, When I provide a complicated question that require a lot long response, what should I do to receive all the information?
Because I only get the first line of the response! thank you

[@computingpower]

Thank you Wassim! Have you checked if the option "see more" Is available?

[@PigtailAIDevelopers]

Thanks a lot! Can you tell me how can I use "gpt-3.5-turbo" as model, when I change to that model throws an error. Thanks again

[@computingpower]

Hello. For using "gpt-3.5-turbo" or "gpt-4" (if OpenAI gave access to you), you must use the Chat Completion API: platform.openai.com/docs/guides/chat
which is different from this one.

[@computingpower]

I forgot to mention that I have a new video using gpt3.5-turbo and Python for an AI Voice assistant project:
ruclips.net/video/UCuaPy45F7I/видео.html

[@PigtailAIDevelopers]

@@computingpower Super!! Thanks dude. I will watch it

[@the_tech_senseii]

And the test?

[@computingpower]

I haven't worked on the tests for this video, but I can do it in another video.

[@JBRandria]

thank you , but when i try , i don't have response in body , just 201 Created as Status

[@computingpower]

Hello, what API call did you try? The one for listing the models or getting an answer for ChatGPT?

[@JBRandria]

@@computingpower Finaly i have to change :
/*const params: CreateCompletionRequest = {
prompt: input.question,
model: input.getModelId(),
temperature: input.getTemperature(),
max_tokens: input.getMaxTokens(),
};*/
to
const params: CreateCompletionRequest = {
prompt: input.question,
model: 'text-davinci-003',
temperature: 0.9,
max_tokens: 2048,
}
there is an error with input.getTemperature(),input.getModelId() and input.getMaxTokens(),