AI infrastructure company Baseten is close to completing a $1.5 billion funding round that would value the business at as much as $13 billion. The deal comes five months after the startup raised $300 million at a $5 billion valuation.
The proposed financing would increase Baseten’s headline valuation by 160% in less than half a year. However, the transaction has not yet been finalised.
Investors Enter at Two Different Valuations
The round uses a dual-tier structure, with some investors purchasing shares at an $11 billion valuation and others investing at $13 billion. Altimeter Capital, Conviction, Spark Capital, Sands Capital, and Wellington Management are reportedly co-leading the deal.
Split-priced rounds allow a company to raise money from different investors under separate terms. They can produce a higher public valuation while giving some participants access at a lower price.
Baseten previously announced a $300 million Series E in January. IVP and CapitalG led that financing, with participation from investors including Nvidia, Altimeter, Battery Ventures, BOND, Greylock, and Blackbird Ventures.
That round followed a $150 million Series D completed approximately nine months earlier. Baseten has therefore raised increasingly large rounds as investors direct more capital toward companies supporting AI inference.
Baseten Focuses on Running AI Models
Founded in 2019, Baseten provides infrastructure for deploying, operating, and scaling AI models in production. Inference is the process through which a trained model generates an answer or performs a task after receiving a request.
Baseten’s inference platform handles tasks such as GPU scheduling, autoscaling, model optimisation, monitoring, and distributing workloads across cloud providers. It supports open-source, custom, and fine-tuned models.
The company positions its services as an alternative to relying entirely on proprietary model providers such as OpenAI and Anthropic. Customers can use lower-cost open-source models for suitable tasks while continuing to use closed models where stronger capabilities are required.
Baseten says its infrastructure can optimise models for latency, throughput, and cost. Its customers include AI coding company Cursor, recruitment startup Mercor, and medical information platform OpenEvidence.
Demand for inference services is increasing as more companies move AI applications from testing into regular use. Unlike model training, which occurs before deployment, inference costs continue for every query, generated image, or completed task.
The proposed funding round would provide Baseten with additional capital to secure computing capacity, improve its software, and compete with other inference providers. The company has not disclosed when it expects to close the financing.
Featured image credits: Magnific.com
For more stories like it, click the +Follow button at the top of this page to follow us.