What is Ollama?
Ollama is a tool for building with open models through local execution or cloud access. Ollama serves individuals handling day-to-day tasks, power users running multiple agents, and teams scaling their use of open models. Local model usage on a user's own hardware is unlimited, while cloud requests are billed using model-specific token rates.
Ollama states that prompt and response data is never logged or used for training. Cloud models and compute resources are hosted primarily in the United States, with requests potentially routed to Europe for additional capacity. Ollama also requires hosting partners to maintain no-logging, no-training, and zero-data-retention policies.
What Ollama can do
Ollama combines open-model access with cloud usage management and team administration. Available capabilities depend on the selected model and plan: tool calling applies to cloud models trained to support tools, while shared billing and administrative controls are associated with team plans. Enterprise adds controls over model availability and spending.
- Ollama runs open models locally or in the cloud.
- Ollama tests supported cloud models for tool calling and real agent workflows.
- Ollama measures cloud usage through input, cached input, and output tokens.
- Ollama Team shares usage credits across unlimited users.
- Ollama Enterprise restricts team access to specific models.
- Ollama Enterprise sets cost budgets for users and API keys.
How Ollama works
Ollama offers local execution on users' own hardware and cloud access through subscription-free or subscription-based plans. Cloud requests consume usage credits at the selected model's rates. Included plan credits are used before purchased credits, and Team requests draw from a shared organizational balance. The pricing page links to a list of cloud-enabled models.
Ollama applies concurrency limits to cloud requests. Pay as you go and Pro include 3 concurrent requests, while Max and Team include 10. Additional requests enter a queue until capacity becomes available. Requests are rejected when that queue is full, so concurrent agent workloads need to account for the selected plan's limits.
Who Ollama is for
Ollama addresses several usage patterns through its plan descriptions. Pro is positioned for shorter, well-defined daily tasks, while Max targets power users running multiple agents simultaneously. Team serves organizations needing shared usage, billing, and administration. Enterprise is intended for larger teams seeking volume usage pricing and additional access controls.
- Ollama supports individuals running open models on their own hardware.
- Ollama supports power users working with multiple agents.
- Ollama supports teams pooling cloud usage credits.
- Ollama supports organizations managing model access and spending.
Ollama pricing and plans
Ollama Pay as you go has a $0 subscription price, no service fees, and per-token cloud charges. Pro costs $20 per month or $200 per year and includes $60 in monthly usage credits. Max costs $100 per month and includes $300 in monthly usage credits, access to larger pro models, and early access to new models.
Ollama Team is listed as early access at $500 per month, with unlimited users and $1,000 in shared monthly usage credits. Enterprise uses custom pricing. Included credits reset monthly and do not roll over, including on annual plans. Every plan can purchase usage credits, and running models on personal hardware remains unlimited.
Ollama strengths and limitations
Ollama combines local model execution, cloud access without a subscription, and paid plans for individual and organizational workloads. Privacy commitments cover prompt and response data, while Team and Enterprise provide billing and administrative capabilities. Cloud performance varies with model size, architecture, and hardware optimization rather than following a single stated speed guarantee.
- Ollama cloud usage remains subject to model-specific token charges.
- Ollama subscription credits expire at the monthly reset rather than rolling over.
- Ollama queues or rejects requests beyond available concurrency capacity.
- Ollama lists fast mode as coming soon.
- Ollama lists shared projects, skills, and instructions as coming soon.
Ollama FAQ
What is Ollama?
Ollama is a tool for building with open models locally or in the cloud. Ollama supports individual tasks, agent workflows, and team model usage, with unlimited execution on personal hardware and cloud access billed through usage credits.
Is Ollama free?
Ollama offers a $0 subscription Pay as you go plan that includes local model execution, but cloud requests are charged per token. Running models on personal hardware is unlimited. Paid Pro, Max, and Team subscriptions include monthly usage credits.
Does Ollama store prompts or use them for training?
Ollama states that prompt and response data is never logged or used for training. Ollama also requires cloud hosting partners to follow no-logging, no-training, and zero-data-retention policies when hosting open models.
Does Ollama support tool calling?
Ollama supports tool calling for cloud models trained to use tools. Ollama tests those models for tool calling and real agent workflows before release; support depends on the model rather than applying to every cloud model.
How many concurrent requests does Ollama allow?
Ollama allows 3 concurrent requests on Pay as you go and Pro, and 10 on Max and Team. Additional requests are queued until a slot opens, while requests arriving when the queue is full are rejected.
Do Ollama usage credits roll over?
Ollama included monthly usage credits do not roll over. Pro, Max, and Team credits refresh on the subscription's monthly reset date, including for annual plans. Included credits are consumed before any purchased usage credit balance.
From the community
No reviews yet
Be the first to review this tool. What you write helps the next person choose.
