Ollama Plans Pricing
Ollama is free and open-source for local inference (no auth, no charge, no rate limit on http://localhost:11434). Ollama Cloud is a hosted-inference add-on with three tiers (Free, Pro, Max), measured by GPU utilization rather than tokens. Cloud usage resets on a 5-hour session and 7-day weekly cycle. All tiers permit unlimited use of open models on the user's own hardware; tier differences apply only to ollama.com cloud-model concurrency and weekly cloud usage.
Ollama Plans Pricing is the machine-readable pricing-plan profile for Ollama on the APIs.io network, in the API Commons Plans shape.
It defines 4 plans, covering freemium and subscription tiers, with named plans including Local, Cloud — Free, Cloud — Pro, Cloud — Max.
This profile was written by API Evangelist and has not been verified against a page the provider published; treat every figure as unconfirmed until a dated transcription replaces it.
Tagged areas include Artificial Intelligence, Large Language Models, and Models.
generated, dated 2026-05-05.
Compare with the market: Artificial Intelligence pricing
Plans
Run Ollama on your own hardware. Open source. No account or API key required for the local server at http://localhost:11434.
- Unlimited local inference
- Open-source models
- No authentication required for localhost
- Ollama API
- Ollama OpenAI Compatibility API
- Ollama Anthropic Compatibility API
Free hosted inference at ollama.com. Requires an ollama.com account and API key. Cloud usage is metered by GPU time, not tokens.
- Run cloud models from CLI/API
- Unlimited public models
- Run models on your own hardware
- Ollama Cloud API
$20/month or $200/year. Run 3 cloud models at a time; 50x more cloud usage than Free.
- 3 concurrent cloud models
- 50x cloud usage of Free
- Ollama Cloud API
$100/month. Run 10 cloud models at a time; 5x more usage than Pro.
- 10 concurrent cloud models
- 5x cloud usage of Pro
- Ollama Cloud API
Sources
Work with this as data
Every plan here is available over the APIs.io API and to AI agents over MCP.