Wan2.5 Text to Video

The Wan2.5 Text-to-Video workflow allows you to generate high-fidelity, audio-visual synchronized dynamic videos directly from natural language descriptions. Powered by AI multimodal technology, it natively supports various sound types (human voice, ASMR, sound effects, music, etc.) and multiple languages, offering seamless audio-visual sync, flexible resolution options, and precise instruction adherence. Create videos up to 10 seconds long with rich details and strong dynamic expressiveness—perfect for diverse storytelling and creative applications. With a built-in prompt optimization switch, you can easily enhance output quality, making it an ideal tool for high-quality AI video creation.

The Wan2.5_Text2Video node is not free to use. Each 5-second video generation (any resolution) costs 300 credits; for 10 seconds, the cost doubles to 600 credits.