Ollama
Run large language models on your own computer
- Category
- Coding & Development
- Pricing
- Local use is free with no service fees. Cloud plans: Free ($0, starter credits), Pro $20/mo or $200/yr ($60 usage credits), Max $100/mo ($300 credits), Team $500/mo (early access, $1,000 shared credits), Enterprise custom. Cloud models are also pay-as-you-go per million tokens.
- Best for
- Who it is for: developers building scripts, agents or apps on open models, teams that need prompts to stay on their own hardware, and hobbyists with a decent GPU or Apple Silicon Mac.
- Official site
- ollama.com
- Last checked
- 2026-10-20
Ollama is a great choice for developers and tinkerers who want to run open-source models on their own machine with no service fees and private prompts. Skip it if you need frontier-quality models with no hardware limits and no tuning, or want a polished chat product out of the box. It handles downloading, running and serving open models behind a simple CLI and local API on macOS, Windows and Linux, and plenty of third-party tools plug into that API. For models too large for your machine it also offers a cloud option with monthly plans that include usage credits, and pay-as-you-go token pricing beyond that.
Who it is for: developers building scripts, agents or apps on open models, teams that need prompts to stay on their own hardware, and hobbyists with a decent GPU or Apple Silicon Mac. Who it is not for: people with weak hardware who want large models at fixed cost (cloud use is metered), non-technical users wanting a ready-made chat app, or teams needing a managed enterprise inference SLA from the free tier.
Key features
One-command model management
Pull and run supported open models with a single command.
Local-first inference
Models run on your hardware by default, with unlimited use and no service fees.
CLI and REST API
A local API makes it easy to connect scripts, editors and agent frameworks.
Cloud models
Larger open models run on Ollama's infrastructure, billed through plan credits or per million tokens, with lower off-peak rates.
Tool calling
Supported on models trained for it, including cloud models.
Modelfiles
Define custom system prompts and parameters for model variants.
Cross-platform
Runs on macOS, Windows and Linux.
Pricing breakdown
Free
- Unlimited local use on your own hardware
- Starter credits for starter cloud models
- 1 concurrent cloud request
Pro
- $60 in monthly usage credits
- 3 concurrent requests
- Larger cloud models
- Run multiple models at once
Max
- $300 in monthly usage credits
- 10 concurrent requests
- Early access to newest models
Team
- $1,000 in shared usage credits
- Unlimited users
- Centralized billing and admin
- Priority support
Enterprise
- Volume usage pricing
- Model access controls and budgets
- Dedicated support
Pros and cons
Pros
- Local inference has no service fees and no usage cap beyond your hardware
- Setup takes minutes and works the same on macOS, Windows and Linux
- Prompts stay on your device in local mode, and the site says prompts are never stored or trained on
- Many third-party UIs, editors and agent frameworks already integrate with its API
- Cloud option lets you reach bigger models without buying a GPU, with transparent per-token rates (for example gpt-oss:20b at $0.07 input / $0.30 output per million tokens)
Cons
- Which models are practical depends on your RAM and GPU; large models can be slow or impossible locally
- Cloud usage is metered: included credits do not roll over, and extra credits are bought separately
- Concurrency is capped (Free 1, Pro 3, Max 10); extra requests queue or are rejected when the queue is full
- Team is listed at $500/mo as early access, a big jump from individual plans
- Mostly CLI and API; you will likely add a separate chat front end
- Peak-hour cloud rates are higher (weekdays 12:00-18:00 UTC)
Alternatives to Ollama
Frequently asked questions
Is Ollama free?
Running models on your own hardware is free and unlimited. The Free cloud tier gives starter credits for starter models; larger cloud models need a paid plan or purchased credits.
What does Ollama Pro cost and include?
$20/mo or $200/yr, with $60 in monthly usage credits and 3 concurrent requests. Unused credits do not roll over.
What hardware do I need?
It runs on macOS, Windows and Linux. Small models run on modest machines, but bigger models need substantial RAM or a capable GPU; I did not verify exact requirements.
Is my data private?
Local mode keeps prompts on your machine, and the site says prompts are never stored or trained on.
How do I use it with other tools?
Through its local REST API, which many editors, chat UIs and agent frameworks support.
Ready to try Ollama?
Head to the official site to explore pricing and start a free trial where available.
Visit Ollama →