Ollama
A command-line tool and local server for running open-weight language models on your own Mac, Windows or Linux machine, with a simple API for other apps.
What is Ollama?
Ollama downloads open-weight models such as Gemma, Qwen, DeepSeek or gpt-oss from its library and runs them on your own computer. You chat with a model in the terminal or the desktop app, and other programs talk to it through a local REST API, which is why many chat interfaces and coding agents list Ollama as a backend. It uses llama.cpp as one of its engines.
Good fit if
- You want to use language models privately, without sending prompts to a cloud API.
- You are building an app or agent and want a local model server that installs in one step and keeps a stable API.
- You plan to put a chat interface such as Open WebUI or a coding agent in front of it.
Look elsewhere if
- You need a complete chat app for a team, with accounts, documents and sharing: Ollama is the engine, so pair it with Open WebUI or AnythingLLM.
- You serve many users at once on GPU servers: vLLM is built for high-throughput serving.
Before you adopt it
Ollama is MIT-licensed, with no paid parts in the repository. It can also run some models in Ollama's cloud; those run on Ollama's servers, not yours, so check which kind you are using if privacy is why you chose it. Large models need plenty of memory: beyond small models, plan on a recent GPU or an Apple Silicon machine.
Drafted with AI from the project's README and repository data, then checked against that data on 30 October 2026. Report a correction
Ollama health: is it still maintained?
89 of 100, actively developed
Read from GitHub on 1 November 2026. How the score works
Ollama license: is it open source?
Ollama uses MIT, an open-source license approved by the OSI.
- June 2023First published on GitHub.
- TodayMIT.
How to run Ollama yourself
- Docker
- Dockerfile or compose file in the repository
- Compose file
- None in the repository
- Setup effort
- Not rated: no compose file to judge from
- Kubernetes
- No Helm chart in the repository
- Desktop and mobile
- Builds for macOS, Windows, Linux, including ARM in GitHub releases
- Interface languages
- English only
Install instructions: ollama.com or the README on GitHub.
Alternatives to Ollama
- vLLMA high-throughput and memory-efficient inference and serving engine for LLMs 84 health · Apache-2.0 · 96k stars
- KoboldCppKoboldCpp - Run GGUF AI models locally with a KoboldAI UI. One File. Zero Install. 93 health · AGPL-3.0 · 12.1k stars
- FastGPTFastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box… 82 health · FastGPT Open Source License · 30k stars
- Open WebUIUser-friendly AI Interface (Supports Ollama, OpenAI API, ...) 76 health · Open WebUI License · 157k stars
- Cherry StudioYour AI workspace for models, agents, and productivity. 81 health · AGPL-3.0 · 53.2k stars
Questions about Ollama
Is Ollama open source?
Yes. Ollama is published under MIT, an OSI-approved open-source license.
Is Ollama still maintained?
Its health score is 89 of 100 (actively developed). It had 367 commits in the last 90 days, the latest on 31 October 2026, and its latest release came out on 28 October 2026.
How do I run Ollama?
You can run it with Docker, as an app for macOS, Windows, Linux, including ARM. The README on GitHub has the exact steps.
What are the alternatives to Ollama?
Close alternatives in the AI chat & local LLMs category include vLLM, KoboldCpp, FastGPT.