Ollama, the Palo Alto-based company behind one of the most popular open-source tools for running large language models locally, announced on July 9 that it has raised a $65 million Series B round led by Theory Ventures. The financing brings the company's total funding to $88 million and lands as Ollama's platform crosses 8.9 million monthly active developers.
Key Highlights
- $65 million Series B led by Theory Ventures, with participation from Benchmark, 8VC, Y Combinator, Pace Capital, 49 Palms, GTMFund, and additional angel investors.
- Total capital raised now stands at $88 million.
- The platform reports 8.9 million monthly active developers, roughly double its January 2026 figure.
- More than 67,000 community-built integrations and partnerships with every major model lab and hardware vendor.
- Founders Jeffrey Morgan and Michael Chiang declined to disclose revenue or post-money valuation.
Details
Founded by Jeffrey Morgan and Michael Chiang, Ollama lets developers bundle, manage, and run open-weight large language models directly on their own machines — whether Mac, Windows, or Linux — rather than sending prompts to cloud APIs. The project has become a default entry point for local AI, gathering more than 176,000 stars on GitHub and a sprawling ecosystem of community integrations.
The company said it will use the new capital to invest in its product, grow its open-source developer community, scale its cloud compute footprint, and bring on key hires. Morgan and Chiang would not discuss revenue or valuation with reporters.
Impact
Ollama's growth reflects a broader shift toward running capable open-weight models on local hardware, a trend accelerated by increasingly strong open models from labs across the industry. For developers wary of per-token cloud pricing, data-privacy constraints, or vendor lock-in, running models locally has moved from a niche experiment to a mainstream workflow.
The company has also expanded beyond pure local execution. Ollama Cloud now routes select models — tagged with a :cloud suffix — to Ollama's own servers, billing inference by GPU time rather than per token. A free tier offers GPU time, with paid subscriptions reported up to $100 per month, giving developers a path to run larger models than their local hardware can handle without abandoning the same familiar interface.
Background
Ollama emerged from Y Combinator and grew primarily through developer word-of-mouth rather than paid marketing. Its command-line simplicity — pulling and running a model with a single command — helped it become shorthand for local AI in much the same way earlier tools became shorthand for containers or version control. In June 2026 the project added ollama launch, a single command that spins up a full coding agent with environment variables configured and the model downloaded automatically if missing.
What's Next
With fresh capital and a rapidly expanding user base, Ollama is positioning itself as more than a local runtime — evolving into a hybrid platform that spans on-device inference and cloud compute. The open question the funding does not answer is how the company will convert nearly 9 million developers into durable revenue, a challenge every open-source infrastructure business eventually faces. For the MENA developer community, where bandwidth costs, data residency, and offline capability often matter, tools that keep AI workloads local remain especially relevant.
Source: TechCrunch