Seedance2.5 1080P: Text to Video
Generate cinematic 1080P video clips from text descriptions using Seedance 2.5, supporting clips up to 30 seconds with synchronized sound, music, and dialogue across any aspect ratio. Ideal for short comedy scenes, product showcases, and creative storytelling.
简介
This workflow turns plain text descriptions into cinematic 1080p video clips with optional synchronized sound using Seedance 2.5. At its core is the ByteDance2TextToVideoNode, which sends your prompt and generation settings (duration, resolution, aspect ratio, FPS, seed) to the Seedance 2.5 backend and returns a finished video stream that can include music, ambience, and simple dialogue aligned to the scene. The SaveVideo node then writes the result to disk as an MP4, preserving the audio track when present.
你需要提供
- 正在读取工作流
在 ComfyUI 中运行。 此工作流无法在本页面运行。你可以在 Comfy Cloud 中打开（已预先加载所需内容），或下载后在自己的 ComfyUI 中使用。
在 Cloud 中试用常见问答
Include audio cues directly in your prompt (e.g., “uplifting synth music,” “footsteps echo,” or “character whispers: ‘Hello’”) and ensure audio is enabled in ByteDance2TextToVideoNode. Keep cues short and specific so Seedance 2.5 can align them plausibly to the scene.
This workflow is designed for clips up to 30 seconds. 1080p (1920x1080) is the default, but you can set other aspect ratios by adjusting width/height (e.g., 1080x1920 for vertical or 1080x1080 for square). Higher resolutions or longer durations may increase generation time and are not guaranteed beyond the stated limits.
Using the same prompt, settings, and seed often yields similar results, but exact reproducibility is not guaranteed across API versions or model updates. For stable iterations, keep all parameters identical and only change one variable at a time.
SaveVideo writes an MP4 file to your configured output directory (commonly ComfyUI/output). If the ByteDance2TextToVideoNode returns an audio track, it is preserved in the saved MP4.