Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Issues with Blackbox Provider #2433

Open
DeveleplorAbdullahH opened this issue Nov 26, 2024 · 4 comments
Open

Issues with Blackbox Provider #2433

DeveleplorAbdullahH opened this issue Nov 26, 2024 · 4 comments
Assignees
Labels
bug Something isn't working

Comments

@DeveleplorAbdullahH
Copy link

There are so many issues with Blackbox provider. It can't read the system prompt and gpt-4o model can not answer. Maybe, there are issues with the servers.

@DeveleplorAbdullahH DeveleplorAbdullahH added the bug Something isn't working label Nov 26, 2024
@Linden10
Copy link

I asked this same question and the answer is here:
#2360 (comment)

@Linden10 Hi, Blackbox is sometimes overloaded and because of this the dialogue, instructions or system settings are lost. For this provider such data losses are quite normal already. That's why I didn't add this provider to the bug list.

I tested this manually with Blackbox in g4f gui, sending messages one at a time and testing the system prompt and found out that Blackbox randomly breaks, ignoring the system prompt and such after responding due to the reasons above.

Yup! Nothing that can be done about it sadly...

@DeveleplorAbdullahH
Copy link
Author

Now, it means that it can't be fixed?

@DeveleplorAbdullahH
Copy link
Author

And there is an issue with Copilot Provider. I wrote a long system prompt to Copilot provider and when I ask something to Copilot provider it says: "RuntimeError: Error: {'event': 'error', 'id': '0', 'errorCode': 'text-too-long'}"

@TheFirstNoob
Copy link

@DeveleplorAbdullahH Hi. Regarding the issue with blackbox. No, g4f can't fix it because the problem is on the provider's side. Technically, you can implement a Json or similar file where you will locally store instructions, dialogue, etc. and let the AI ​​process it before each request, but this will greatly increase the load on the provider.

Copilot error says that you have exceeded the context limit on input parameters. Try to reduce them. The size of the limits is set by the providers themselves, as well as the models used.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

4 participants