# n8n The community package is `@apergrex/n8n-nodes-franklab`, version `0.2.4`. It contains 25 FrankLab nodes. [npm package](https://www.npmjs.com/package/@apergrex/n8n-nodes-franklab) · [source and node documentation](https://github.com/d906090-rgb/n8n-nodes-franklab). ## Install In your self-hosted n8n, sign in as owner or administrator. Open **Settings → Community Nodes → Install**, enter `@apergrex/n8n-nodes-franklab@0.2.4`, review the community-package notice and install. Follow [n8n's installation guide](https://docs.n8n.io/integrations/community-nodes/installation-and-management/gui-installation) if this menu is unavailable. ## Connect Create FrankLab credentials with your API key. For this package version, use `https://franklab.ru/franklab/api` or `https://franklab.ru` as the base URL; the package restricts accepted hosts. Keep credentials in n8n, not in chat or exported workflows. ## First generation 1. Follow the [balance and price checks](/en/docs) before running a paid operation. 2. Add **Manual Trigger → FrankLab MARS**, operation **Text to Video**. 3. Enter a text prompt, set duration to 5 seconds and aspect ratio to 16:9. Leave Model blank and set Additional JSON to the following: ```json {"model_name":"mars-v1-5","mode":"std"} ``` 4. Run creation once after checking the estimate. Preserve the returned task ID. 5. Use **Get Text Video Status** to poll that same task until completion, then use its video URL. Importing or installing does not generate a video. Creating a task is paid. Do not automatically retry task creation after an ambiguous timeout; inspect the existing task first. A successful estimate does not guarantee provider availability.