Ollama
VerifiedGet up and running with large language models locally.
Description
Ollama is the runtime that made local models mainstream: one command pulls and runs Llama, Gemma, Qwen, DeepSeek and hundreds of other open models on macOS, Windows and Linux, with GPU acceleration and an OpenAI-compatible API that most self-hosted chat UIs build on.
Key features
One-line model pulls from a curated library. OpenAI-compatible REST API. GPU acceleration on Apple Silicon, NVIDIA and AMD. Modelfiles for custom system prompts and parameters. The de-facto backend for Open WebUI and most self-hosted chat frontends.
Who it's for
Developers and self-hosters who want to run open models on their own hardware.
Related products
Explore more tools for similar tasks.
LM Studio
Discover, download and run local LLMs from your desktop.
FreeView productAgent Names by username.md
Give your AI agent an identity it can own.
PaidView product
Shipvela
Website hosting from GitHub, your CLI or your coding agent.
FreemiumView productGPT4All
Run local LLMs privately on everyday desktops.
Free to self-hostView product
orke
AI coding agents, terminal tasks, project web views, and Git in one workspace
FreeView productKitmorrow
127 tools, including AI Rewrite. One membership.
PaidView product
Building an AI product too?
Paste your product URL, review the autofilled details and submit for approval.