Rate Limits

The API enforces rate limits per API key using a token-bucket algorithm. Limits vary by subscription tier.

---

Limits by tier

Tier	Plan	Per-key limit	Per-account limit
0	Free	60 req/min	100 req/min
1	Indie	300 req/min	500 req/min
2	Platform	1,000 req/min	2,000 req/min
3	Scale	3,000 req/min	6,000 req/min

Rate limits are enforced at the edge (Cloudflare Durable Objects) — low latency, no central bottleneck.

---

Rate limit headers

Every response includes:

X-RateLimit-Limit: 300
X-RateLimit-Remaining: 298

When a limit is exceeded, the response also includes:

Retry-After: 4

Retry-After is the number of seconds to wait before retrying.

---

Handling 429s

TypeScript

// The SDK handles 429s automatically with exponential backoff + jitter.
// Default: 3 retries, respects Retry-After header.

// To customize:
const client = new Xavigate({
  apiKey: 'test_sk_YOUR_KEY',
  maxRetries: 5,
});

Python

# The SDK handles 429s automatically with exponential backoff + jitter.
client = xavigate.Xavigate(
    api_key="test_sk_YOUR_KEY",
    max_retries=5,
)

Manual retry

# Read Retry-After from the 429 response header:
# Retry-After: 4
# Wait that many seconds, then retry the identical request.
sleep 4
curl -X POST https://api.xavigate.com/v1/nature/subjects ...

---

Best practices

• Use the SDK. Auto-retry with jitter is built in — you don't need to implement backoff yourself.
• Batch lazily. Don't fan out N API calls simultaneously. Process items in sequence or small batches with brief pauses.
• Monitor X-RateLimit-Remaining. If it approaches 0, slow down before you hit a 429.
• Need higher limits? Upgrade your tier in the dashboard under Billing.