Ollama

Ollama is a platform for accessing open language models for inference, either on local hardware or through cloud-hosted models. It addresses the need to run open models with control over where inference happens, with the company stating that prompts and responses are not logged or trained on. Models can be run locally without usage limits, or accessed through cloud infrastructure hosted primarily in the United States, with routing to Europe and Singapore for additional capacity.

Cloud models are accessed through usage credits, with pricing set per million tokens for input, cached input, and output, and off-peak rates available outside certain weekday hours and on weekends. Models support tool calling, and cloud models are tested with real agent workflows before release. The platform offers a Free plan with starter usage and access to starter models, Pro and Max plans with monthly usage credits, concurrent request limits, and access to larger models, and a Team plan with shared usage, centralized billing, and administration. An Enterprise tier adds model access controls, cost budgets, and dedicated support. Running models on personal hardware remains unlimited regardless of plan.