# Rate Limiting Speechace is architected to be highly available and auto-scalable to serve request spikes as needed. The API provides a concurrency guarantee of up to 16 concurrent Transactions per Second (TPS). Above that limit, the API will throttle calls returning `error_too_many_requests.` A higher concurrency limit is available with custom plans and annual contracts. You may contact us at to request a higher limit.