Popular open source AI tool Ollama has raised a $65 million Series B, led by Theory Ventures, founder and CEO Jeff Morgan tells TechCrunch.

This round follows an earlier $15 million Series A led by Benchmark’s Peter Fenton. In total, the company has raised $88 million.

Ollama, which launched in 2023, helps developers run open-ended AI models on their PCs, getting them up and running in minutes. It has been praised by developers on countless training sites, videos, blogs, and social media posts. It has amassed 176,000 stars and nearly 17,000 forks on GitHub.

Developers can also use Ollama to find models and access larger, more complex models hosted on its neocloud through various subscription levels, from free to $100 per month. It also tracks usage based on GPU time, not token limits.

If the mission of helping developers build more easily on their PCs sounds vaguely familiar, it should. Morgan and his co-founder Michael Chiang previously helped build Docker Desktop. They landed on Docker after it bought their previous startup, Kitematic. Docker creates containers that help cloud applications move easily from one cloud to another, or from desktop to cloud, abstracting away all the annoying hardware configuration issues.

So Ollama essentially did for AI what Docker and Docker Desktop did for the cloud.

“Open models started appearing in 2023, but they were very difficult to use,” Morgan said. At the time they were aimed at researchers, not programmers. “As a result, it was very difficult to get them up and running.” Three years after its launch, Ollama is now “used by more than 8.9 million developers every month, occupying 85% of Fortune 500 companies and growing like crazy,” he said. All with only 14 employees.

That professional experience is what attracted Benchmark’s Peter Fenton to lead its previous round and join the board.

“What Jeff and Michael built with Docker is used by over 10 million developers every day. The creative powers to create a product that reaches ubiquity for developers is extremely rare,” Fenton told TechCrunch.

Morgan and Fenton declined to discuss the startup’s revenue and its new valuation. However, Morgan says the proving point for Ollama as a business occurred around January, when OpenClaw became popular. That’s when larger open models “suddenly became capable of doing these agency tasks, like coding. Obviously, we saw the explosion of assistants like OpenClaw and this idea that open models can do real work.”

Since then, the industry has been excited by the idea that paying users (particularly deep-pocketed enterprises and fast-growing AI application layer startups) will increasingly turn to more affordable open models, reserving their use of closed models like Anthropic for more cases as needed.

“I still think this is the part where most of the debate gets it wrong. It’s not an either/or question,” Fenton says of open versus closed AI models. There will be a lot of business for both, he says. However, every company with high inference expenses (the costs of using models) has a “vital existential project” that pushes it to move “to open models,” he says.

There is plenty of evidence that these startups and companies are already turning to open models for their daily needs. That obviously bodes well for Ollama’s cloud business.

But what’s even more interesting is that Ollama is another example of how AI is spawning a huge new crop of open source projects that are becoming companies pursued by venture capitalists. There are open source inference providers such as Inferact, creator of vLLM, and RadixArk, creator of SGLang. There is OpenClaw and its alternatives such as NanoClaw. There are even small startups building their own open models from scratch, like Arcee.

To be sure, not all Ollama fans have been happy that the company has sought to make a living. About a year ago, a bunch of blog and social media posts complained that their cloud business was diverting attention from their beloved free project and cited Ollama as an example of the so-called “enshittification” of development tools, as the trend is called.

But Morgan sees its cloud service as an evolution of its open source mission to help programmers easily find and use models. Those large, cutting-edge, open models are often “too big to run on your own computer. So we said, ‘Hey, let’s help find the computer for that,'” he explained.

Adds board member Fenton: “Nothing has changed for the core product that is free on the desktop. There is no change to the premise that this is the place where you can discover and run local models.”