Baseten is the fastest way to ship AI-native products and apps that are fast, reliable, and cost-efficient at scale. Powered by the Baseten Inference Stack, which serves GenAI models with optimized, modality-specific Model Runtimes and Multi-cloud Capacity Management.

Thanks Ryan! We’re really excited to launch BaseTen today to ease the pains of building applications powered by machine learning. The idea of BaseTen came from pains we felt ourselves, as machine learning engineers, being stuck with our models in Jupyter notebooks waiting for DevOps and product engineers to help us operationalize them. We’ve built a product that allows users to seamlessly integrate ML models with their custom business logic, expose them as APIs, and ultimately assemble user-facing apps that utilize those APIs. It’s been quite a journey to get here — we’ve been working on BaseTen for almost 18 months now and iterating with users to make sure we were building something that was 1) simple, 2) powerful, and 3) a joy to use. Our team is listening to this thread and would love your feedback. And we’re happy to answer any questions.

Quite excited for Baseten to be launching today! We've had the chance to explore the product at Primer, and it's remarkable the leverage they create for teams. It's plausible that Baseten will save us thousands of hours of time in the next few years.

I've worked with the founding team before. A+++. Excited for this!