infobro
Writing & Content53Image58Video28Audio & Voice33Chat & Assistants24Productivity46Coding & Development41Agents & Automation28Marketing & SEO21Sales & Support45Business & Finance96Operations21Data & Research23Education25Design22Health & Life38
O

Ollama

Run open AI models locally or in the cloud on macOS, Windows and Linux

Free · from $16.67/mo ollama.ai
Visit website

Own Ollama? Claim it, free to correct the facts and get the verified badge.

ollama.ai
Ollama screenshot

Pricing

Free · from $16.67/mo

API

Yes

Open source

Yes

Company

Ollama

GitHub stars

185K

Platforms

macOSWindowsLinuxAPI

What is Ollama?

Ollama runs open AI models on your own computer or on Ollama's cloud. It works on macOS, Windows and Linux, and you can use models such as DeepSeek, Gemma, GLM, gpt-oss, Kimi, MiniMax, Mistral and Nemotron.

Running models on your own hardware is unlimited and costs nothing. Cloud models are billed per million tokens using usage credits. Some models get cheaper off-peak rates. Paid plans include a monthly credit allowance, access to larger models and more concurrent requests. Ollama says it never logs or trains on prompt or response data, and it requires zero data retention from the hosting partners it works with.

Ollama suits developers who build with open models, people who want private local inference, and teams that need shared billing and admin controls. The project is on GitHub, which lists about 184.8K stars.

Key features

  • Run open models locally with unlimited usage
  • Cloud-hosted open models with pay-as-you-go token pricing
  • Off-peak discounted rates for some models
  • Tool calling support on compatible cloud models
  • Concurrent requests up to 10 on higher plans
  • API keys with per-key cost budgets (Enterprise)
  • Shared team billing and administration
  • Usage alert email at 90% of monthly allowance

Use cases

Who it's for

DevelopersAI engineersPrivacy-conscious usersEngineering teamsEnterprises

Private local inference

Run open models on your own machine so prompts never leave your hardware, with no usage limits.

Building agents with open models

Use cloud models that support tool calling, with concurrent requests for running several agents.

Accessing large models without GPUs

Use larger cloud models such as Kimi or DeepSeek Pro when local hardware isn't enough.

Team model usage

Share credits, billing and administration across unlimited users on the Team plan.

Ollama pricing

From the vendor's site, checked Oct 26, 2026; we re-check prices monthly. Prices change — confirm on their pricing page.

PlanFreeProMaxTeam (early access)Enterprise
Paid monthlyFree$20/mo$100/mo$500/moCustom quote
Paid yearlyFree$16.67/mo−17%———
Monthly usage creditsStarter credits$60$300$1,000 shared—
Concurrent requests1310——
Model accessStarter modelsLarger pro modelsEarly access to newest models—Access controls
Priority or dedicated support———Priority supportPrivate Slack channel
Also
  • Run models locally
  • No service fees
——
  • Centralized billing
  • Cost budgets for users and API keys

What the free plan gets you

  • Pay as you go: $0 subscription, pay per token for cloud models
  • 3 concurrent requests
  • Running models on your own hardware is unlimited
Cost for a team, per month1 person5 people20 people
Proflat, up to 1$16.67——
Maxflat, up to 1$100——
Team (early access)flat$500$500$500

Billed yearly where the plan offers it. — means the plan doesn't allow that many people.

Which plan do you need?

  1. MistralPro$14.99/mo
  2. OllamaPro$16.67/mo
  3. LM StudioBionic+$20/mo
  4. GPT4AllIndividual$20/mo

The cheapest paid plan with a published price that fits. Custom and sales-quoted plans aren't included.

Get an email when Ollama changes its prices

One mail per change, a stop link in each. Nothing else.

How Ollama compares

  • At $16.67/mo, Ollama starts cheaper than 2 of 3 similar tools with published prices (median $18.34).
  • Ollama has a free plan, like 3 of the 4 alternatives below.
  • It is open source, as is 1 other.
ToolStarts atFree planFree trialAPIOpen source
OOllama$16.67/mo—
LLM Studio$20/mo——Ollama vs LM Studio
GGPT4All$20/mo———Ollama vs GPT4All
RReplicate————Ollama vs Replicate
MMistral$14.99/mo—Ollama vs Mistral
Compare 8 capabilities
CapabilityOllamaLM StudioGPT4AllMistral
Works fully offline—
Optional cloud models—
OpenAI-compatible API
Apple Silicon acceleration——
Runs on CPU only——
Fine-tuning———
Speaker diarization———
Team admin / SSO

Capabilities as stated on each vendor's own site; — means the site doesn't say. The closest alternatives in our directory by category. “Starts at” is the cheapest paid plan per month, as each vendor lists it.

Ollama data and privacy

Trains AI on your content

No

“We do not use your inputs or outputs to train any AI models”

Data retention

Cloud prompts/responses processed transiently, not stored beyond request time

“When using cloud-hosted models, we process this content transiently to provide the Service and this content is not stored beyond the time required to fulfill the request.”

Where data is stored

Primarily United States; may route to Europe

“Ollama hosts models and compute resources primarily in the United States. To serve global demand, we may route to Europe for additional capacity.”

Security and compliance

None stated

How Ollama compares on data training. From Ollama's own privacy and security pages , checked Oct 28, 2026. We only show what the vendor states in writing; “not stated” means the pages don't say.

Pros and cons

Strengths

  • Local model use is free and unlimited
  • Prompt and response data is never logged or trained on
  • Open source project on GitHub
  • Published per-token prices for every cloud model

Limitations

  • Free cloud usage is limited to starter models unless you buy credits
  • Unused monthly credits do not roll over
  • Free plan allows only 1 concurrent cloud request
  • Cloud hosting is primarily in the US

Ollama FAQ

Is Ollama free?

Yes. Running models locally is free and unlimited. The Free plan also includes starter cloud credits. Paid plans start at $20/mo, or $16.67/mo billed annually.

Which platforms does it support?

macOS, Windows and Linux.

Is my data used for training?

No. Ollama says prompt and response data is never logged or trained on, and hosting partners must have zero data retention policies.

Do unused credits roll over?

No. Included credits refresh at each monthly reset.

Where are cloud models hosted?

Primarily in the United States. Requests may be routed to Europe for extra capacity.

Used Ollama? Rate it

Sign in to rate this tool and share how you use it.

Ollama in our articles

Ollama

Free · from $16.67/mo

Try Ollama

Tools similar to Ollama

D
DeepSeek

AI chat assistant and low-cost LLM API for consumers and developers