Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Local runtime - Ollama - verified

Ollama: install, API, platforms and features

Ollama is a local LLM runtime. As its docs describe it: Developers who want to download and run open models from the command line, with a local REST API on port 11434 and built-in launchers for coding agents such as Claude Code and Codex. The latest release is v0.40.6, published October 16, 2026. License: MIT (open source).

Homepage - Source on GitHub - MIT

Latest release
v0.40.6
Released
2026-10-16
License
Open source
OpenAI API
Yes

Install

Copied verbatim from the page linked under each command. Labels are worded as the source words them.

macOS

Terminal
curl -fsSL https://ollama.com/install.sh | sh

Source: github.com/ollama/ollama

Windows

Terminal
irm https://ollama.com/install.ps1 | iex

Source: github.com/ollama/ollama

Linux

Terminal
curl -fsSL https://ollama.com/install.sh | sh

Source: docs.ollama.com/linux

Docker (CPU only)

Terminal
docker run -d -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama

Source: docs.ollama.com/docker

Run a model

The first-run example from the official docs. OpenAI-compatible endpoint: http://localhost:11434/v1

Terminal
ollama run gemma4

Source: github.com/ollama/ollama

Features

What the Ollama docs say it supports. Not stated means the docs do not say either way.

Ollama documented features, each with its source
FeatureValueSource
InterfaceCLI, Server, Desktop app, DockerSee sources below
PlatformsmacOS, Windows, Linuxgithub.com/ollama/ollama
BackendsCPU, CUDA, ROCm, Metal, Vulkandocs.ollama.com/gpu
Model formatsGGUF, Safetensorsdocs.ollama.com/import
OpenAI-compatible APIYeshttp://localhost:11434/v1docs.ollama.com/api/openai-compatibility
Anthropic-compatible APIYesdocs.ollama.com/api/anthropic-compatibility
Tool callingYesdocs.ollama.com/capabilities/tool-calling
Structured outputsYesdocs.ollama.com/capabilities/structured-outputs
Vision (image input)Yesdocs.ollama.com/capabilities/vision
Multi-GPUYesdocs.ollama.com/faq
MCP clientNot statedDocs do not say
LicenseMITOpen sourcegithub.com/ollama/ollama/blob/main/LICENSE
Latest releasev0.40.6October 16, 2026github.com/ollama/ollama/releases/tag/v0.40.6

Models to run with Ollama

36 of the open-weight models in our local models directory have a tag in the Ollama library:

Compare Ollama

FAQ

Does Ollama have an OpenAI-compatible API?

Yes. Endpoint, per its docs: http://localhost:11434/v1. Source: https://docs.ollama.com/api/openai-compatibility

Does Ollama have an Anthropic-compatible API?

Yes, per https://docs.ollama.com/api/anthropic-compatibility

Which platforms and GPUs does Ollama support?

Its docs list macOS, Windows and Linux, with these backends: CPU, CUDA, ROCm, Metal and Vulkan.

How do I install Ollama?

On macOS: "curl -fsSL https://ollama.com/install.sh | sh", as shown at https://github.com/ollama/ollama. This page lists 4 install commands: macOS, Windows, Linux and Docker (CPU only).

Is Ollama open source?

Yes. It is released under MIT; the license is at https://github.com/ollama/ollama/blob/main/LICENSE.

What is the latest version of Ollama?

v0.40.6, published October 16, 2026 (https://github.com/ollama/ollama/releases/tag/v0.40.6), as of October 18, 2026.

Which model formats does Ollama run?

GGUF and Safetensors, per https://docs.ollama.com/import

Sources

Last verified from the pages below.

Newsletter

Local AI, every week

One email per week. New model releases, runtime updates and what they mean for what you build.

Keep exploring

Open-weight models you can run

Download sizes, licenses and the run command for each model.

All local LLM runtimes

Every runtime in one feature matrix, with sources.

Picking a runtime for coding agents

Ollama, LM Studio, vLLM and llama.cpp for agent workloads.

Setup guides

Point your coding tools at the model you picked.