
Baseten
baseten.co · Inference & Hosting
Production inference platform for deploying open and custom models with dedicated, autoscaling GPUs.
Baseten is a production inference platform. It offers Model APIs (ready to use, optimized open models billed per million tokens), dedicated deployments of open, fine tuned or custom models on autoscaling GPUs, and training. Its Truss framework packages a model and its serving code for deployment.
Pricing, per the official pricing page: the Basic plan is $0 a month and pay as you go for dedicated deployments, Model APIs and training. Pro adds priority access to high demand GPUs and higher Model API rate limits, and Enterprise adds self hosted or hybrid deployment in your own VPC, custom SLAs and data residency controls. Baseten is SOC 2 Type II and HIPAA compliant. In the Vercel AI SDK it is available through @ai-sdk/baseten. Alternatives include Modal, Together AI, Fireworks AI and Replicate.
Baseten is one of 10 inference and hosting tools on Drydock. See all inference and hosting tools or how we check facts.