faq
Answers to common questions and solutions to frequent issues.
Jump to:
getting_started
Do I need API keys to use x47?
No! x47 works out of the box without any setup. You can add your own API keys later if you want higher rate limits or access to specific provider accounts. See our API keys guide for details.
Is x47 free to use?
Yes, x47 is free to use immediately without any configuration. If you choose to add your own API keys, you'll pay for usage directly to the AI providers based on their pricing.
Can I use x47 offline?
The x47 interface can load offline after your first visit, but you need an internet connection to execute prompts since they require API calls to the AI providers.
custom_api_keys_(optional)
Why would I add my own API keys?
Adding your own keys gives you higher rate limits, access to your existing provider accounts and billing, and potentially access to models not available by default.
Where do I configure API keys?
Click the 'api_keys' button in the console header. See our detailed API keys guide for step-by-step setup instructions for each provider.
Are my API keys secure?
Your keys are stored in your browser's localStorage and only sent directly to AI providers when running prompts. Never sent to x47 servers.
My API key isn't working
Check our API keys troubleshooting guide for common issues like incorrect copying, expired keys, or billing problems.
models_&_outputs
Why do different models give different outputs?
Each AI model is trained differently and has unique characteristics. GPT, Claude, Gemini, and others have different training data, architectures, and fine-tuning, leading to varied responses. This is exactly why x47 exists - to help you compare these differences.
Why is a model taking so long to respond?
Response times vary based on: 1) Model complexity (larger models are slower), 2) Response length requested, 3) Provider server load, 4) Your internet connection. Reasoning models like DeepSeek R1 intentionally take longer as they 'think' through problems.
Can I stop a running prompt?
Currently, once a prompt is sent to the API, it cannot be cancelled. The models will complete their responses. You can close the tab, but you'll still be charged for the API usage.
Why are my outputs being cut off?
Increase the 'Response Length' setting. The default 'medium' setting limits responses to ~512 tokens. For longer content, use 'long' or 'very_long'. Note that longer responses cost more.
variables
Why isn't my variable being detected?
Variables must use the exact syntax $variable_name with only letters, numbers, and underscores. Common issues: using spaces ($my variable), starting with numbers ($1variable), or using special characters ($my-variable).
Can I use the same variable multiple times?
Yes! If you use $topic three times in your prompt, all instances will be replaced with the same value you enter in the variables panel.
How do I include a literal $ in my prompt?
Currently, there's no escape sequence. As a workaround, use a variable like $dollar and set its value to '$', or rephrase to avoid the dollar sign.
data_&_privacy
Where is my data stored?
All data (tabs, history, API keys, settings) is stored in your browser's localStorage. Nothing is stored on x47 servers. This means your data doesn't sync across devices or browsers.
How do I transfer my data to another browser?
Export your history from the history sidebar, then import the JSON file in the new browser. For API keys and tabs, you'll need to set them up again manually.
Will clearing my browser data delete my x47 data?
Yes. Clearing site data, cookies, or localStorage for x47.ai will delete all saved tabs, history, API keys, and settings. Export your history first if you want to keep it.
common_errors
"Invalid API key" error
Double-check that you copied the entire key without extra spaces or characters. Some keys are quite long. Try generating a new key from your provider's dashboard.
"Rate limit exceeded" error
You've hit the provider's rate limit. Wait a few minutes before trying again. Consider upgrading your API tier for higher limits, or reduce the number of simultaneous model calls.
"Insufficient quota" or "billing" error
Your API account needs payment information or additional credits. Visit your provider's billing dashboard to add funds or set up payment.
"Context length exceeded" error
Your prompt (including system prompt and conversation history) is too long for the model. Try: 1) Shortening your prompt, 2) Using a model with a larger context window, 3) Breaking your task into a prompt chain with smaller steps.
Model output is empty or just whitespace
This can happen with very short prompts or unusual formatting. Try: 1) Adding more context to your prompt, 2) Adjusting the creativity setting, 3) Changing the output format setting.
still_need_help
Can't find an answer to your question?
Check the documentation
Browse our detailed guides for API keys, settings, variables, and prompt chains.
Send feedback
Use the feedback button in the console to report bugs or suggest improvements. We read every submission.
tips_for_reporting_issues
When reporting an issue, include:
- •What you were trying to do
- •What happened vs. what you expected
- •Any error messages (exact text or screenshot)
- •Your browser name and version
- •Which models you were using