A step by step guide on installing Ollama and Testing It with Postman

Windows

Download the Windows installer for Ollama from the official website and run the setup program. After installation, open Command Prompt or PowerShell and confirm it works by typing ollama --version. Next, download a model such as llama3 by running:

ollama pull llama3

This command downloads the model so it can run locally on your machine. Ollama automatically starts a local API service on http://localhost:11434. Install Postman from the Postman website and open the application. Create a new POST request to http://localhost:11434/api/generate. Add the header Content-Type: application/json. In the body, use JSON like this:

{
 "model":"llama3",
 "prompt":"Explain REST APIs",
 "stream":false
}

Click Send. If the response includes generated text, Ollama and the API connection are working successfully.

Ubuntu

On Ubuntu, install Ollama using the official installation script. Open a terminal and run:

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

After installation, confirm it works using ollama --version. Download a model using ollama pull llama3. This step may take several minutes depending on your internet speed because the model files are large. Ollama automatically exposes a REST API on port 11434. Install Postman on Ubuntu by downloading the Linux package or using Snap. Create a POST request to the same endpoint http://localhost:11434/api/generate with a JSON request body containing a prompt. When Postman returns generated text from the model, the installation and API communication are verified.