Baseten is closing a $1.5 billion funding round as the AI infrastructure startup bets on cheaper alternatives to dominant players like OpenAI and Anthropic. The round uses a dual-tiered structure with investors split between an $11 billion and $13 billion valuation. The company provides software and computing capacity specifically for businesses using lower-cost AI models.

Tech executives have predicted that inference—the computing-intensive process of generating AI responses—will consume the entire AI industry's budget this year. This shift is already materializing as enterprises face mounting costs for closed-source models. Tuhin Srivastava, Baseten's co-founder and CEO, notes that major AI labs bundle inference infrastructure for free with their models, creating pricing pressure for alternatives.

The quality gap between open-source and proprietary models has narrowed significantly over the past few years. Large enterprises and startups are now pushing back against escalating AI expenses, driving demand for open-source options that users can download and modify at no cost. This cost-conscious trend represents a fundamental market shift away from premium AI services.

Baseten's fundraising signals growing investor confidence in the inference infrastructure market, potentially reshaping how businesses approach AI deployment costs. The dual valuation structure suggests strong institutional backing for this alternative model strategy.