Overview
How might we design rate limit warnings so people can trust and act on AI output?
When this pattern fits
- Essential for applications using AI APIs with rate limits, developer tools, and platforms where proactive limit management prevents service interruptions.
When to skip or lighten it
- Truly unlimited internal deployments where limits never apply.
- Background batch jobs better served by queues than interactive warnings.
- When limits are so high that constant nags train users to ignore them.
States
State model coming soon
Key UX elements
Key UX elements coming soon
Anti-patterns to avoid
Failing only after submit with a cryptic HTTP 429 and no retry guidance.
Warnings that push upgrade without showing remaining allowance.
Inconsistent units between the warning and the billing page.
Blocking the composer with no estimate of when capacity returns.
How products use it
| Product | Implementation |
|---|---|
| ChatGPT / Claude Pro | Usage and limit messaging when approaching plan caps. |
| Cursor | Premium model and agent usage warnings against plan quotas. |
| API consoles (OpenAI, Anthropic) | Rate limit headers and dashboard alerts for RPM/TPM. |
| Perplexity | Pro/feature gates when free limits are exhausted. |
Implementation
Copy this prompt to generate a production-ready implementation in Cursor, Claude Code, Lovable, or any AI coding agent.
Generate a production-ready implementation of the "Rate Limit Warnings" AI interface design pattern.
Pattern Definition:Frequently asked questions
What are rate limit warnings in AI UX?
Rate limit warnings tell users they are nearing API or plan quotas before a hard failure, and explain wait, upgrade, or model-switch options.
When should a rate limit warning appear?
Show a soft warning as the user approaches the ceiling (for example under 20% remaining) and a hard, actionable state at the limit, not only a raw error after send.
How do rate limits relate to cost transparency?
Cost transparency shows price per action. Rate limits show throughput or plan caps. Users need both when agents can burn quota quickly.