Modal Auto Endpoints: Optimized inference you own

Modal has launched Auto Endpoints, a self-serve platform that enables teams to deploy and own production-grade LLM inference with a single command while maintaining full visibility into the underlying code, metrics, and infrastructure. Unlike traditional managed inference providers that hide implementation details, Modal Auto Endpoints give users complete control and transparency over their inference stack, from GPU selection to engine optimization, without requiring them to manually manage the entire serving infrastructure. The service is built on Modal's AI infrastructure platform and supports open models like GLM 5.2 with pay-as-you-go pricing and automatic scaling.