Configure LLM

Locimens agents call models through a local LLM gateway (port 3210). You choose a provider in Settings → AI. Keys are encrypted on disk by the gateway — not stored in browser localStorage.

Choose a provider

Ollama setup

1. 1

   Install Ollama

   Download from ollama.com and install on the same PC as Locimens.

2. 2

   Pull a model

   In a terminal run: ollama pull llama3.2 (or your preferred model).

3. 3

   Confirm Ollama is running

   Default endpoint is http://127.0.0.1:11434. Ollama should be running before agents start.

4. 4

   Select Ollama in Locimens

   Settings → AI → Provider → Ollama. The app will fetch and list your available Ollama models automatically; pick one from the dropdown.

Custom / remote LLM settings

Locimens supports more than two providers. If your provider exposes an OpenAI-style HTTPS API (for example, a hosted provider, a company gateway, or your own server running vLLM / LM Studio / etc.), use the Custom provider. If you run Ollama on another machine (home server / private VPS), you can also point Locimens at that remote Ollama endpoint — your Ollama models will still be listed automatically.

Custom provider (OpenAI-compatible API)

The Custom provider is for any HTTPS endpoint that implements OpenAI-style chat completions: `POST …/v1/chat/completions`. You provide the API base URL, API key, and the exact model id your server expects.

1. 1

   Open Settings → AI

   In the Locimens desktop app, open Settings and go to the AI section.

2. 2

   Enter your API base URL

   Paste the base URL (example format: https://api.example.com/v1). Locimens will call `…/v1/chat/completions` under the hood.

3. 3

   Enter your API key

   Paste the key issued by your provider (or your gateway). Store it only on devices you control.

4. 4

   Enter your model id

   Type the exact model id string your provider expects (example: gpt-4o-mini). For Custom providers, the model list may not be auto-discovered — use the provider’s docs to find the correct id.

5. 5

   Test connection, then Save

   Use Test connection to validate URL/key/model. Save when it succeeds, then select “Custom” as your active provider.

OpenRouter setup

1. 1

   Create an OpenRouter key

   Sign up at openrouter.ai, create an API key, and add billing credits if required.

2. 2

   Paste the key in Locimens

   Settings → AI → Provider → OpenRouter → enter API key → Save. The gateway encrypts and stores it locally.

3. 3

   Select a model

   Choose from the dynamic model list (fetched automatically). Agents and Telegram sync to the selected model.

Running agents 24/7

For agents to run 24/7, the Locimens desktop app and your MT5 terminal must be running continuously on a machine that stays online. Many users do this on a dedicated PC or a private server/VPS they control.