API Reference

Error Handling

Understand error codes and how to handle them gracefully.

Error Response Format

All errors return a consistent JSON structure:

JSON

{
  "error": {
    "message": "Human-readable error description",
    "type": "error_type",
    "code": "specific_error_code"
  }
}

Error Codes

400invalid_request_error

The request body is malformed or missing required fields.

Examples: Invalid JSON, missing 'messages' field, invalid parameters

401invalid_api_key

The API key is missing, invalid, or revoked.

Examples: Missing Authorization header, malformed key, disabled key

402payment_required

Account has insufficient credits or no active subscription.

Examples: Zero balance, expired subscription

403forbidden

Your plan doesn't have access to the requested resource.

Examples: Free tier attempting to use infe-titan model

404model_not_found

The requested model does not exist.

Examples: Typo in model name, deprecated model

429rate_limit_exceeded

You've exceeded your rate limit for requests or tokens.

Examples: Too many requests per minute

500api_error

An unexpected error occurred on our servers.

Examples: Internal server error, temporary outage

502api_error

The upstream inference provider returned an error.

Examples: Provider timeout, service unavailable

Handling Errors

Best practices for error handling:

from openai import OpenAI, APIError, RateLimitError, AuthenticationError
client = OpenAI(
    api_key="your-infe-api-key",
    base_url="https://api.infe.io/v1"
)
try:
    response = client.chat.completions.create(
        model="infe-pulse",
        messages=[{"role": "user", "content": "Hello!"}]
    )
except AuthenticationError:
    print("Invalid API key. Check your credentials.")
except RateLimitError:
    print("Rate limit exceeded. Wait and retry.")
except APIError as e:
    print(f"API error: {e.message}")

Retry Strategy

For transient errors (429, 500, 502), implement exponential backoff:

Code	Should Retry?	Recommendation
400	No	Fix the request
401	No	Check API key
402	No	Top up credits
403	No	Upgrade plan
429	Yes	Wait 1s, then 2s, then 4s...
500	Yes	Retry up to 3 times
502	Yes	Retry with backoff

Troubleshooting

"Invalid API key"

Ensure the Authorization header is Bearer YOUR_KEY
Check for extra spaces or newlines in the key
Verify the key hasn't been revoked in your dashboard

"Insufficient credits"

Check your balance in the dashboard
Top up or upgrade your plan
Reduce token usage with shorter prompts

"Model not found"

Use exact model IDs: infe-pulse, infe-titan
Check available models with GET /v1/models
Some models require higher tiers

Need Help?

If you're encountering persistent errors, include the x-request-id header from the response when contacting support.

Contact Support →