Chat Overview
Active

Ollama

Run models locally or use optional cloud models through the same model tooling.

  • Local models with a command line and API
  • Optional cloud models for larger workloads
  • Local hardware costs differ from cloud usage credits

Ollama downloads and runs supported models on a local computer. It also offers cloud models, making it possible to use models that exceed the machine's available memory.

Local or cloud

Local execution uses the computer's processor, memory, and storage. After a model has been downloaded, local inference does not require a hosted model endpoint. Cloud models run remotely and require account access. Select the model and execution location explicitly when handling sensitive material.

How pricing works

Local inference does not incur an Ollama cloud inference bill, but hardware and power still have a cost. Cloud use consumes usage credits; subscriptions include different allowances. Downloaded models retain their own license terms. For a desktop interface with local model management, see LM Studio .

Sources

Checked .

From the archive