Ollama, the open source tool that lets developers run open-weight AI models locally on their own machines, has raised a $65 million Series B round led by Theory Ventures. The funding, which follows a $15 million Series A led by Benchmark's Peter Fenton and brings total investment to $88 million, underscores growing investor appetite for open source AI infrastructure as enterprises look to cut inference costs by adopting cheaper open models alongside closed ones such as Anthropic's.

Launched in 2023 by Jeff Morgan and Michael Chiang — both formerly involved in building Docker Desktop — Ollama says it is now used by more than 8.9 million developers each month, sits within 85% of the Fortune 500, and has gathered 176,000 stars and nearly 17,000 forks on GitHub, all with just 14 employees. Beyond the free desktop product, the company runs a "neocloud" hosting larger models via subscription tiers from free to $100 a month, billed by GPU time rather than tokens. Morgan and Fenton declined to reveal revenue or valuation, and though some fans have criticised the cloud push as "enshittification" of a beloved free tool, both insist the core desktop product remains free and unchanged.