Setting Up Ollama

Snipset includes powerful AI capabilities, such as AI Chat, AI Snippet Generation, and Semantic Search (OmniSearch). To use these features securely and privately, Snipset relies on Ollama, a tool that runs AI models on your computer or through Ollama Cloud for hybrid models.

Step 1: Install Ollama (Easiest Method)

1. Download the Installer Go to the official Ollama Download Page and click on the "Download for Windows" button to get the standard .exe installer.
2. Run the Installer Double-click the downloaded OllamaSetup.exe file and follow the standard on-screen instructions. This process is automatic and takes just a few moments.

Step 2: Verify Ollama is Running

Ollama runs silently in the background on your machine.

• Open your web browser and navigate to http://localhost:11434
• If Ollama is running correctly, you will see a simple white page that says: "Ollama is running".

Ollama Tips: Downloading Models Using PowerShell

Now that Ollama is installed, you need to "pull" (download) the AI models using the PowerShell terminal. Follow these strict step-by-step instructions:

1. Open PowerShell Press the Windows key on your keyboard, type PowerShell, and press Enter to open the terminal window.
2. Sign In to Ollama The recommended chat model runs on Ollama Cloud, so sign in once. Type this command, press Enter, and finish the sign-in in the browser window that opens (create a free Ollama account if you do not have one):
   ollama signin
3. Download the Chat Model Type the pull command for the recommended chat model and press Enter:
   ollama pull gemma4:31b-cloud
   This is Snipset's default chat model, Gemma 4 31B on Ollama Cloud. The download is under 1 KB and it puts no load on your RAM or GPU, but it needs internet. If it is unavailable, pull the fallback minimax-m2.7:cloud instead.
4. Download the Embedding Model Next, type this command and press Enter. This downloads the 229 MB embedding model required for Semantic Search:
   ollama pull hf.co/lmstudio-community/embeddinggemma-300m-qat-GGUF
   Wait again for the "success" confirmation.
5. Verify Your Models Finally, type this command and press Enter to list your downloaded models and ensure they are ready to use:
   ollama list
   You should see gemma4:31b-cloud and hf.co/lmstudio-community/embeddinggemma-300m-qat-GGUF listed.

Step 3: Connect Snipset to Ollama

1. Go to Snipset → Settings → AI & Models → AI Configuration.
2. Ensure Base URL is http://localhost:11434 (or http://127.0.0.1:11434; both are accepted, and the app default is http://127.0.0.1:11434).
3. Look for the green Online indicator next to the URL.
4. Select your Chat Model (gemma4:31b-cloud, recommended; or llama3.2:3b, a 2.0 GB local model, for fully offline chat) and Embed Model (hf.co/lmstudio-community/embeddinggemma-300m-qat-GGUF) from the dropdowns.

Troubleshooting