Ollama logo
AI Tool Profile

Ollama

Ollama downloads and runs open AI models on your own macOS, Windows or Linux computer, with a command line, local API and desktop app. Local use is free; paid plans cover Ollama's cloud models.

Website
ollama.com
Pricing model
Open Source
Price start
Free

Description of Ollama

Ollama is software for running open AI models on your own computer or, optionally, in Ollama's cloud. Its homepage describes it as the easiest way to run and build with open models locally or in the cloud, and says prompts are not stored or used for training. Installers exist for macOS, Windows and Linux, and an official Docker image is published. The code is MIT licensed on GitHub.

Ollama screenshot
Ollama official website

You download a model by name, then chat with it from the command line or the desktop app, or call it from other software through a local API. Ollama's documentation says its API can be used with OpenAI-compatible and Anthropic-compatible clients, and it offers Python and JavaScript libraries. The GitHub page lists integrations with coding tools such as Claude Code and Codex, chat interfaces such as Open WebUI and LibreChat, and libraries such as LangChain and LlamaIndex.

Pricing applies only to cloud models. The Free plan is $0, lets you run models locally, includes starter usage credits and one concurrent cloud request. Pro is $20 per month, or $16.67 per month billed annually, with $60 of monthly usage credits, larger models and three concurrent requests. Max is $100 per month with $300 in credits and ten concurrent requests. Team is $500 per month in early access, and Enterprise is custom.

Because models named with a cloud tag run on Ollama's servers, they are not local. A buyer wanting fully offline use should pick models without that tag and check that the download size fits free memory. Ollama's own pages do not state a hardware minimum on the homepage, so check the model page for size before downloading.

Features of Ollama

  1. Runs open models on macOS, Windows and Linux from a command line, a desktop app or a local API.
  2. Free plan covers local use; Pro at $20 a month adds usage credits for Ollama's hosted cloud models.
  3. Works with OpenAI-compatible and Anthropic-compatible clients and lists integrations such as Open WebUI, LibreChat, LangChain and LlamaIndex.

Frequently asked questions

Is Ollama free?

Running models on your own machine is free; the Free plan is $0. Paid plans such as Pro at $20 a month are for credits to use Ollama's cloud models, which are not local.

Which operating systems does Ollama support?

The official GitHub page lists installers for macOS, Windows and Linux, plus an official Docker image. The software is released under the MIT license.

Does Ollama have an API?

Yes. Ollama's documentation links an API reference and says OpenAI-compatible and Anthropic-compatible clients can use it. Python and JavaScript libraries are also provided.

Website: https://ollama.com/

Company: Ollama

Key facts

Free, MIT-licensed runner for open models on macOS, Windows and Linux with a local API. Paid plans start at $20 a month but only cover Ollama's cloud models, which do not run on your machine.

Pricing model Open source
Free tier Yes · Run models locally at no cost; starter cloud credits; 1 concurrent cloud request
Starting price $20 / month
Data residency Not verified
Compliance (as stated by vendor) Not verified
Integrations Claude Code, Codex, Open WebUI, LibreChat, Chatbox, LangChain, LlamaIndex, OpenCode
Public API Yes
Self-hostable Yes
Official site Not verified
Can a general AI assistant replace it? A hosted chatbot needs no setup. Ollama fits when you need offline use, private data on your machine or a local API; cloud-tagged models give that up.
Last verified 2026-10-10 by editor

Alternatives & Similar Tools

Overchat AI logo
Overchat AI Freemium

Overchat AI is an all-in-one AI platform that combines 50+ leading models with 100+ mini apps for chat, image and video generation, document work, writing, and study tasks.

l
llama.cpp Open Source

llama.cpp is a free, MIT-licensed C/C++ engine for running language models on your own hardware, including CPUs and many GPUs. It includes an OpenAI-compatible server and sits under several local AI apps.

Jan logo
Jan Open Source

Jan is a free, open-source ChatGPT alternative from Menlo Research that runs local or online models on Windows, macOS and Linux, with an OpenAI-compatible local API server.

LM Studio Bionic logo

LM Studio, now shown as LM Studio Bionic, runs local language models and an agent that works on documents and code. The free plan covers local models, offline voice transcription and LM Link; Bionic+ is $20 a month.

Mistral Vibe logo

Mistral's Vibe, which Mistral's help center says replaces the Le Chat name, is an AI assistant for research, coding and everyday tasks. A Free plan has limited messages; Pro costs $14.99 a month.

D

DeepSeek is an AI chat assistant from a Chinese research company, offered as a mobile app, web chat and desktop client. The official site calls the app free to use; the API is billed per token.

Update history

  1. 2026-10-10 · ai alternative note updated (editor)
  2. 2026-10-10 · editor verdict updated (editor)
  3. 2026-10-10 · self hostable updated (editor)
  4. 2026-10-10 · api available updated (editor)
  5. 2026-10-10 · integrations updated (editor)
  6. 2026-10-10 · pricing page url updated (editor)
  7. 2026-10-10 · starting price usd updated (editor)
  8. 2026-10-10 · free tier updated (editor)
  9. 2026-10-10 · pricing model updated (editor)

Beyond the tool

Need the outcome rather than the software?