Ollama is a free, open source tool that runs large language models locally with a single command, for example ollama run llama3. It manages model downloads, GPU acceleration and a local REST API that most AI applications can talk to. It started as a command line tool and now also ships a simple desktop app with a chat window on Mac and Windows, and it works with many third-party interfaces such as Open WebUI.

Ollama is for developers, tinkerers and anyone who wants a local model as a backend for other software. If you want a full featured chat interface out of the box, LM Studio is more comfortable, but Ollama is lighter and scripts well.

No install required - run AI models on the cloud

BitVector Prism is the easiest way to start: pick a model, type a prompt and generate in a clean web app, with nothing to configure. BitVector is also available on Discord and on the web (SpyGlass).

Try BitVector

What you need

macOS 12 or later (Apple Silicon recommended), Windows 10 or 11, or Linux. 8 GB RAM for 7B models, 16 GB for 13B, 32 GB for 33B; an NVIDIA or AMD GPU makes generation much faster. Models take 2 to 40 GB each.

How to get and install Ollama

Download Ollama from ollama.com for Mac or Windows and run the installer, or on Linux run the install script from the site.
Open a terminal (or the Ollama app) and type ollama run llama3.2 (or another model name from the library page); the model downloads and a chat prompt appears.
Browse the model library on ollama.com for Qwen, Gemma, Mistral, DeepSeek, Phi and vision models, each with the exact command to run it.
Optional: install a web interface such as Open WebUI, which connects to Ollama and gives you a ChatGPT-style page in your browser.

Download only from the official links on this page. Installers found elsewhere may be out of date or tampered with. ModelTrends.ai hosts no files.

No install required - run AI models on the cloud

PirateDiffusion is Telegram only and built for pros: thousands of models, LoRAs and workflows driven by chat commands, with unlimited generation on a fixed price plan.

Try PirateDiffusion

How to use Ollama

Chat in the terminal or the app window; type /bye to exit a terminal chat.
List downloaded models with ollama list, remove one with ollama rm name, and pull a new one with ollama pull name.
Pick a size tag to match your hardware, for example qwen2.5:7b or qwen2.5:14b; smaller tags run on smaller machines.
Use the API at http://localhost:11434 from your own scripts (it is OpenAI compatible), or point apps like Open WebUI, Continue or Zed at it.
Create a custom model with a Modelfile to set a system prompt and parameters, then ollama create myname -f Modelfile.
Vision models (llava, llama3.2-vision, qwen2.5vl) accept images: drag a file into the chat or pass the path in the terminal.

Tips and common problems

Set OLLAMA_MODELS to move the model folder to a bigger drive.
If the GPU is not used, check that your driver is current; ollama ps shows whether a model runs on GPU or CPU.
Keep the context window (num_ctx) modest unless you need long documents; it has the largest effect on memory.
Ollama can import GGUF files you download yourself through a Modelfile with a FROM line pointing at the file.

Models you can run in Ollama

Ollama runs the open text models under Text / LLM in this catalog that are published in its library or available as GGUF files, which covers Llama, Qwen, Gemma, Mistral, DeepSeek and most others.

Text / LLM Models

No install required - run AI models on the cloud

No GPU, no downloads, no Python. These services run the same open source models on their own servers, so you can start generating from a browser, Discord or Telegram in a minute.

Try BitVector

Try PirateDiffusion

Links

These links go to the makers' own sites. ModelTrends.ai is not affiliated with them and does not host any software.

Other local apps

How to install ComfyUI

The node-based powerhouse for Stable Diffusion, Flux, Wan video and almost every new open source model. Free, open source, for Windows, Mac and Linux.