Troubleshooting common issues
This guide helps you resolve common issues when using Routic. If you encounter an error not covered here, contact support with your request ID.
API Key issues
Key creation fails
Symptom: When creating a new API key in the console, you see "We couldn't create your API key. Please try again later."
Cause: The gateway service is temporarily unavailable or experiencing high load.
Solution:
- Wait a few minutes and try again
- If the issue persists after 10+ minutes, contact support
- Check your account balance — keys may fail if balance is too low
Key not found
Symptom: You see "API key not found" when making API calls.
Cause: The key was revoked, expired, or never existed.
Solution:
- Check that your key starts with
sk-and is 48 characters long - Verify the key exists in your dashboard
- If the key was revoked, create a new one
- Ensure you're using the correct Base URL:
https://api.routic.ai/v1
Authentication errors
For 401 authentication errors during API calls, see HTTP semantics & error payloads.
Service availability issues
Service temporarily unavailable
Symptom: Console or API returns "Service temporarily unavailable. Try again later."
Cause: The gateway or backend services are experiencing issues.
Solution:
- Wait 30 seconds to 1 minute and retry
- Check the status page for known incidents
- If you're making API calls, use exponential backoff (see Error handling & retries)
- Contact support if the issue lasts more than 10 minutes
Model call errors
401 Unauthorized
Your API key is invalid, expired, or missing.
| Check | Action |
|---|---|
| Key format | Must start with sk-, 48 characters |
| Key status | Check dashboard — key may be revoked |
| Header | Add Authorization: Bearer sk-xxx |
See HTTP semantics & error payloads for details.
404 Not Found
The requested resource or model doesn't exist.
| Check | Action |
|---|---|
| Model name | Verify the model exists in Models |
| Endpoint URL | Use /v1/chat/completions, not other paths |
| Base URL | Ensure https://api.routic.ai/v1 |
429 Rate Limit
You've exceeded your rate limit (requests per minute or tokens per minute).
| Check | Action |
|---|---|
| RPM/TPM limits | Default: 100 RPM, 10,000 TPM per key |
| Retry-After header | Wait the specified seconds |
| Multiple keys | Rotate across keys (each has independent limits) |
Do not retry immediately — use exponential backoff. See Error handling & retries.
500 / 503 Server Errors
Temporary server-side issues.
| Status | Meaning | Action |
|---|---|---|
| 500 | Internal server error | Retry with exponential backoff |
| 503 | Upstream unavailable | Wait 30s, then retry |
For retry strategies, see Error handling & retries.
Console issues
Console shows Chinese by default
Symptom: The console displays Chinese text even though you prefer English.
Solution: Click the language switcher in the top-right corner to change to English.
Balance shows zero after payment
Symptom: You completed a payment but balance still shows zero.
Cause: Payment processing delay (typically 1–5 minutes for Stripe).
Solution:
- Wait 5 minutes and refresh
- Check your email for payment confirmation
- If balance still shows zero after 10 minutes, contact support with your payment receipt
When to contact support
Contact support when:
- An error persists for more than 10 minutes despite retries
- Your payment succeeded but balance hasn't updated
- You suspect your account or key was compromised
- You need a feature not documented here
Always include:
- Your account email
- The
request_idfrom the error response - The approximate time the error occurred
- Any relevant screenshots
See SLA & support channels for contact methods.
See also
- HTTP semantics & error payloads — full error code reference
- Error handling & retries — retry strategies
- Authentication & API keys — key management
- SLA & support channels — how to reach us