Ollama promotes open models for security teams and enterprises with zero data retention

Ollama says defenders can run open models on their own machines or on its cloud with zero data retention, keeping sensitive data under their control.

In an October 23, 2026 post, co-founder Michael Chiang argues that defensive work such as exploit analysis and red teaming needs models that closed providers often restrict to special access programs, and that attackers already use open models. These are Ollama's claims and have not been independently verified.

In a November 3, 2026 post, Ollama says it is making open models the default for enterprises worldwide, with full zero data retention and more capable models. The post says open models were first used for sensitive work such as private documents and government data. The post also tags Together AI and two people, Vipul Ved Prakash and J. Morgan, though the reports do not state their roles in the announcement.