MiniMax: Text to Video
Generate high-quality videos directly from text prompts. Explore MiniMax's advanced AI capabilities to create diverse visual narratives with professional CGI effects and stylistic elements to bring your descriptions to life.
About
This ComfyUI tutorial walks you through a minimal, reliable text-to-video pipeline powered by the MiniMax API. The core of the workflow is the MinimaxTextToVideoNode, which sends your text prompt to MiniMax’s text-to-video service and returns a generated video clip. A MarkdownNote node is included in the graph to remind you of the provider’s secure network requirements (localhost access, HTTPS usage, and ensuring you can reach the API). Finally, a SaveVideo node writes the returned clip to disk so you can review and reuse it in your projects.
What you provide
- Reading the workflow
Runs in ComfyUI. This workflow can't run on this page. Open it in Comfy Cloud with everything loaded, or download it for your own ComfyUI.
Try in CloudQuestions & answers
No. The MinimaxTextToVideoNode calls the MiniMax API for generation, so heavy compute happens remotely. Your local machine only needs to run ComfyUI and maintain a network connection.
Common causes are networking or credential issues. Confirm you meet the secure environment requirements (localhost access or HTTPS for web services), your API key/endpoint are correct, and your firewall allows outbound requests to the MiniMax API.
Use any parameters the MinimaxTextToVideoNode exposes in your workflow. If a setting isn’t available in the node UI, consult the MiniMax provider docs for supported options. Some services may also respect prompt cues (e.g., 9:16) but this is not guaranteed.
Write specific prompts that include subject, action, environment, and style cues (e.g., cinematic lighting, stylized CGI). Keep motions simple for short clips, avoid overly rapid scene changes, and iterate in small steps. If the node supports it, adjust duration and other parameters conservatively.