Rate Limits

Concurrency

Each API key can run up to 5 concurrent solves by default. Requests beyond that limit are queued and will resolve once a slot opens.

If you need higher concurrency, contact hello@solver.lol.

Rate limit response

When you exceed the concurrency limit, the request will wait in queue rather than immediately returning a 429. If the queue is full, you'll receive:

{
  "error": "rate_limit_exceeded",
  "message": "Too many concurrent requests. Please retry."
}

Best practices

• Reuse your API key across requests — don't create a new key per request
• For bulk workloads, batch requests and respect the concurrency limit
• Implement exponential backoff on 429 responses