Can't create completion giving null error HTTP 429

[Tc554]

Already checked the HTTP 429 issue but still error

Code:

OpenAiService service = new OpenAiService("token here");
CompletionRequest completionRequest = CompletionRequest.builder()
                .prompt(scanner.next())
                .model("text-davinci-003")
                .echo(true)
                .temperature(0.7)
                .maxTokens(96)
                .topP(1.0)
                .frequencyPenalty(0.0)
                .presencePenalty(0.3)
                .build();
CompletionResult result = service.createCompletion(completionRequest); // <---- null here

[Tc554]

Exception in thread "main" retrofit2.adapter.rxjava2.HttpException: HTTP 429 at retrofit2.adapter.rxjava2.BodyObservable$BodyObserver.onNext(BodyObservable.java:57) at retrofit2.adapter.rxjava2.BodyObservable$BodyObserver.onNext(BodyObservable.java:38) at retrofit2.adapter.rxjava2.CallExecuteObservable.subscribeActual(CallExecuteObservable.java:48) at io.reactivex.Observable.subscribe(Observable.java:10151) at retrofit2.adapter.rxjava2.BodyObservable.subscribeActual(BodyObservable.java:35) at io.reactivex.Observable.subscribe(Observable.java:10151) at io.reactivex.internal.operators.observable.ObservableSingleSingle.subscribeActual(ObservableSingleSingle.java:35) at io.reactivex.Single.subscribe(Single.java:2517) at io.reactivex.Single.blockingGet(Single.java:2001) at com.theokanning.openai.OpenAiService.createCompletion(OpenAiService.java:116) at me.TastyCake.Main.main(Main.java:26)

Process finished with exit code 1

[cryptoapebot]

(Sorry saw you already checked code) HTTP 429 is too many requests. I think basic accounts get 10 requests per minute, but with the popularity of the service, it might be handling it in aggregate.

https://status.openai.com/

Make sure that your license can handle the number of requests, i.e. credit card for overages, using the right license key, etc. Also make sure your code can handle non-200 HTTP result codes.

			if ((response.statusCode() / 100) != 2) {
				System.out.println("response code: " + response.statusCode());
				System.out.println("response body: " + response.body());
			} 

CompletionResult uses annotation lombok @Data Make sure that the latest lombok.jar is correctly configured and on path. I used 1.18.24

Finally, for your scanner.next(), make sure the prompt is properly quoted or even clean it from other characters. Hope any of this helps.

[cryptoapebot]

I copied the @Data tag from the CompletionResult code, but now I'm wondering if the capital T is screwing up the annotation. Nope, it's correct in code. Github comments just though I was referencing a diff account of user "DaTa" by using the at sign.

[Tc554]

thank you for responding I do see some outusage but how can I fix it I just want to generate one response

OpenAiService service = new OpenAiService("token here");
CompletionRequest completionRequest = CompletionRequest.builder()
                .prompt(scanner.next())
                .model("text-davinci-003")
                .echo(true)
                .temperature(0.7)
                .maxTokens(96)
                .topP(1.0)
                .frequencyPenalty(0.0)
                .presencePenalty(0.3)
                .build();
CompletionResult result = service.createCompletion(completionRequest); // <---- null here

this code is on the main method running just one time

[cryptoapebot]

I tried recreating your issue, but I can't, so that means maybe it's an environment issue? Try taking .echo out. Only advice I have is to just compare a curl command with the exact same parameters/prompt and see what it says.

[Magnus1987]

I got the exact same issue. I don't think it has to do with the environment as the token is only a string value that can be assigned directly to OpenAiService.

I tried the following (also with echo() ):

CompletionRequest completionRequest = CompletionRequest.builder()
                .model("ada") //Cheapest option
                .prompt("Somebody once told me the world is gonna roll me")
                .build();

[AlexUrrutia]

same issue here, it doesn't work, usage shows 0%

[martylamb]

I'm seeing the same issue with a brand new token on a brand new account, using the smashmouth sample code. Usage on openai is showing 0 requests, and status page says all systems are operational.

I see that this issue is closed. What is the resolution? I'm using v0.12.0. Thanks.

[martylamb]

Nevermind, sorry for the noise. This is clearly not an issue with the library and not your problem to solve.

But for anyone else who stumbles on this thread, adding a credit card to my new account ended up resolving the issue despite the fact that the free account provides $18 in credit. This seems to have bitten quite a few folks.

[chenkangck50]

你需要开通支付账户

[peri103]

I added a credit card to my new account, but still met the same error. Is there any other solution?

[FuadEfendi]

I have the same issue; newly created account says my phone number is already in use and I cannot use credits (I didn't register there before!), so I was forced to add $10. Now, OpenAiApiExample created downloadable "dancing with frog" image, I can see that I was charged $0.02 for that, but console still shows error 429.

I am using v.0.18.2

[UPDATE]

When I run this example 2nd time, after about 10 minutes, I don't have any issue. So, I assume "billing setup" takes some time at their site.

$0.04 already ;)