OutageRadar

Checking… Checking…

Claude API error 529: overloaded_error

What this Claude API error means, whether it is on Anthropic’s side or yours, and what to do now. All times are UTC.

Is it them or you?

Server-side: Anthropic’s capacity

This error comes from Anthropic’s servers, not from your code. If the status below shows a problem, Anthropic has confirmed it. If it shows Operational, the problem may be short-lived, or the status page may not show it yet; our own check is a second opinion.

Checking…

Claude

Official status: checking…

Our check: checking…

What it means

Anthropic’s docs describe a 529 overloaded_error as the API being temporarily overloaded. It can happen when the API has high traffic across all users, so it is not caused by your request or your API key.

It is different from a 429 rate_limit_error, which is about your own organization’s limits.

Shape of the error body, from Anthropic’s docs (the message text varies):

{
  "type": "error",
  "error": {
    "type": "overloaded_error",
    "message": "..."
  },
  "request_id": "req_..."
}

What to do

  1. Retry with exponential backoff. Anthropic’s official SDK already retries server errors like this one automatically and honors the retry-after header; its max_retries setting controls how many times.
  2. If it keeps failing, check Claude’s status above and the official status page.
  3. If you need answers now, switch to a backup (below) until it clears.

Retry it safely in your own code

You may not need this: the official SDKs retry on their own (see above). If you call the API directly, this helper waits and retries, using the server’s requested delay when you pass one. Treat 529 and other 5xx errors as retryable; never retry 400, 401, 402 or 403 errors.

Python

import random
import time


def with_retries(call, is_retryable, retry_after=lambda err: None, attempts=5):
    """Run call(). On a retryable error, wait and try again, up to attempts tries.
    retry_after(err) returns the server's requested wait in seconds, or None."""
    for attempt in range(attempts):
        try:
            return call()
        except Exception as err:
            if attempt == attempts - 1 or not is_retryable(err):
                raise
            wait = retry_after(err)
            if wait is None:
                wait = min(60, 2 ** attempt) + random.random()  # backoff + jitter
            time.sleep(wait)

JavaScript

async function withRetries(call, isRetryable, retryAfter = () => null, attempts = 5) {
  // retryAfter(err) returns the server's requested wait in seconds, or null.
  for (let attempt = 0; ; attempt++) {
    try {
      return await call();
    } catch (err) {
      if (attempt >= attempts - 1 || !isRetryable(err)) throw err;
      let wait = retryAfter(err);
      if (wait == null) wait = Math.min(60, 2 ** attempt) + Math.random(); // backoff + jitter
      await new Promise((resolve) => setTimeout(resolve, wait * 1000));
    }
  }
}

If Claude is down: switch to a backup

These AI APIs work with the official OpenAI library. Change the address and the key, pick a model from that provider’s list, and you’re running again. Each shows its live status, so you don’t switch to one that’s also down.

Test a backup in development before routing production traffic. Pick a model from a different company than the one that’s down. Your code talks to the provider directly; OutageRadar never sees your key. For a full automatic fallback setup, see TokenMargin.

Recent Claude incidents

History logged by OutageRadar.

Full history: Is Claude down?

Source: Claude API errors (Anthropic docs). Last checked against official docs: Nov 10, 2026. If the docs change, they win.