The provider Ollama announced a Series B funding round of $65 million on July 9, 2026. As Ollama states in its blog, the investment firm Theory Ventures is leading the round. For the company, founded in 2023, whose software makes open AI models runnable locally on one’s own computer, this is the second major funding round after a $15 million Series A led by Benchmark partner Peter Fenton.
The Funding Round in Detail
With the fresh $65 million, Ollama says it now has total funding of $88 million since its founding. In addition to Theory Ventures, according to reporting by TechCrunch, Benchmark, 8VC, and Y Combinator are also participating again in the round; as further backers, SiliconANGLE additionally names Pace Capital, 49 Palms, GTMFund, and several angel investors. No official valuation for the round has been disclosed.
Ollama was founded by CEO Jeffrey Morgan and Michael Chiang. Both had previously built the tool Kitematic, which became the basis for Docker Desktop after being acquired by Docker – a track record that carries weight with investors.
From Command-Line Tool to Developer Standard
Ollama positions itself as easy access to open, freely available AI models: instead of complex setup, a single command is usually enough to download and run a model locally. According to company figures picked up by TechCrunch, around 8.9 million developers now use the software monthly – roughly twice as many as in early 2026, when the company had about 4.45 million active users. On GitHub, the project claims more than 176,000 stars and nearly 17,000 forks; SiliconANGLE puts the number of integrations with other developer tools at more than 67,000. Ollama also says 85 percent of Fortune 500 companies use the software in some form – a figure that cannot be independently verified.
Co-founder Morgan has attributed the growth, per SiliconANGLE, to the fact that open models have existed since 2023 but were long too complicated for many developers to set up – the exact hurdle Ollama says it wants to remove. Despite the rapid growth, the team remains comparatively small, at around 14 employees according to the company.
Free Locally, Paid in the Cloud
Ollama’s desktop application for running models locally is free. For models that exceed a computer’s local capacity, the company also operates a cloud service with an identical interface. Billing is based on GPU time rather than token limits, with subscription tiers ranging from free to $100 per month. TheNextWeb reports that monthly token volume through the cloud service has roughly doubled every few months on average – growth that investors such as Benchmark partner Peter Fenton point to as evidence that a rising share of AI usage will run on open rather than closed models going forward.
Assessment: A Bet on Open Models
The round fits a broader pattern of investors putting money not just into the large model providers themselves, but increasingly into the infrastructure surrounding open AI models. Ollama does not act as a model developer but as a neutral layer through which models from different providers – for instance from the open ecosystems around Meta, Alibaba, or DeepSeek – can be run in a unified way.
Whether the rapid user growth can translate into sustainable, recurring revenue through the cloud service remains an open question; Ollama does not disclose specific revenue figures. Still, for companies that prefer to self-host AI models for cost or data-protection reasons rather than sourcing them through large providers’ APIs, Ollama positions itself as a natural point of contact – a market that tends to grow as inference costs for closed models rise.
