Ollama Turbo
9 months ago
- #Privacy
- #AI
- #Hardware
- TurboPreview offers faster model inference with datacenter-grade hardware.
- Allows running larger models by upgrading to newer hardware.
- Ensures privacy and security by not retaining user data.
- Saves battery life by offloading model processing from local devices.
- Turbo is a new way to run open models efficiently.
- Available models during preview include gpt-oss-20b and gpt-oss-120b.
- Compatible with Ollama's CLI, API, and JavaScript/Python libraries.
- No data logging or retention in Turbo mode.
- Hardware is located in the United States.
- Usage limits are in place with future metered pricing planned.