AI model aggregation platform OpenRouter has officially launched its Batch API, offering a 50% discount on per-token prices for both input and output across more than 70 supported models. The service targets non-urgent tasks with all results guaranteed within 24 hours. During a two-week testing period, the platform completed over 230,000 batch tasks with a median completion time of seven minutes. Ninety percent of these tasks were processed within one hour, demonstrating significant throughput capacity for asynchronous AI workloads.