Generate text-to-video clips with optional audio-driven lip-sync via RunComfy.
npx skills add https://github.com/runcomfy-com/skills --skill wan-2-7
Please help me install this Agent Skill. Skill: wan-2-7 Source: https://github.com/runcomfy-com/skills/tree/main/wan-2-7 Command: npx skills add https://github.com/runcomfy-com/skills --skill wan-2-7
Generate high-quality text-to-video clips with motion that matches the prompt, including audio-driven lip-sync when you provide a voice track.
audio_url to drive speaking motions and timing in the generated video.negative_prompt.Ask for a Wan 2.7 talking-head video and include your audio track URL so the result is lip-synced to your voice.
High-intent search queries and answers about installing and using this skill.
To generate a lip-sync video, provide a text prompt and an audio_url containing a 3–30 second WAV or MP3 file. The Wan 2.7 model processes these inputs to create a video where the subject's speaking motions match your supplied voice track.
Lip-sync video generation supports WAV and MP3 audio files. The audio track must be between 3 and 30 seconds long and cannot exceed 15MB in size to be processed by the model.
Yes, you can control the aspect ratio, resolution, and duration of your text-to-video clips. These configurable parameters allow you to tailor the output dimensions to fit specific cinematic or spokesperson video requirements.
Prompt expansion improves text-to-video consistency by automatically rewriting your input prompt. Toggling this feature on helps the model better understand the intent, resulting in generated motion that more accurately matches your requested scenario.
The best way to exclude unwanted elements is by using the negative_prompt parameter. Supplying targeted exclusions prevents the model from generating specific objects, styles, or motions you do not want in the final video.
Use Wan 2.7 when you explicitly need Wan 2.7 features like audio-driven lip-sync or prompt-controlled cinematic motion. The system routes requests to alternative models like HappyHorse 1.0, Seedance 2.0 Pro, Kling Video O1, or LTX 2 based on your specific video generation intent.
agentspace-soCreate avatar videos from audio or scripts via RunComfy model routing.
nagasriramnaniGenerate AI images, videos, and retrieve stock photos from text prompts.
prime-skillsGenerate text-to-video clips with Wan 2.7 via the RunComfy Model API.
0xslineGenerate AI videos via Seedance 2.0, Kling, and MiniMax Hailuo models.