Rate Limits - What to Know About Your AI Provider

When using your own API key with Websito, you're subject to rate limits imposed by the provider you connected (e.g. OpenAI, Anthropic, Groq, etc). If you’re suddenly seeing failed generations or long response times, it might be due to hitting these limits.

⚠️ What Are Rate Limits?

Rate limits control how many requests you can send to your AI provider over a set time period. If you exceed these, your request will be rejected or delayed.

For example:

• OpenAI might limit you to 20 requests per minute
• Anthropic may allow up to 100K tokens per day
• Groq and others may impose concurrent request limits

🚨 Signs You've Hit a Rate Limit

• The AI widget stops responding or gives errors
• Page generation fails repeatedly
• Long delays after submitting a prompt
• You see “rate limit exceeded” or “429” errors in logs or browser console

🔍 How to Check Your Limits

Each provider has its own way to view usage and rate limits:

🛠️ How to Avoid Rate Limit Issues

• Use slower typing or spacing between prompts if generating large content
• Choose less busy hours when demand is lower (for shared services)
• Avoid submitting too many prompts at once
• Use multiple keys (rotate between providers if needed)

🧩 Tip: Use Groq or OpenRouter for High-Speed Requests

Some providers like Groq and OpenRouter have extremely fast processing and generous limits. They’re great for bulk page generation.

🔁 Can I Change My Limits?

Most providers let you request higher limits via their support or billing portal — especially if you're on a paid plan.

✅ Summary

Rate limits are normal and expected when using your own API keys. By knowing your provider's restrictions, you can plan accordingly and avoid delays in your website generation.

If you're unsure, try switching providers in Websito’s AI settings or contact us for suggestions!