qianwen-video-generation
Generate videos through multiple input modes—text descriptions, single images, frame transitions, character role-play, or video editing—powered by Qianwen's Wan models. All operations run asynchronously; submit your request and poll for completion. The skill auto-detects your task and routes to the right model, with detailed reference guides for prompt engineering, polling patterns, and media workflows.
Qianwen Video Generation creates videos from text prompts, images, or frame pairs using Wan models with asynchronous processing.
AI-generated summary based on this skill's SKILL.md
Install
QianWen-AI/qianwen-ai/qianwen-video-generation · repository language: Python
git clone https://github.com/QianWen-AI/qianwen-ai
cp -r qianwen-ai/skills/video/qianwen-video-generation ~/.claude/skills/qianwen-video-generationnpx skillfed install QianWen-AI/qianwen-ai/qianwen-video-generationFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
How do I generate video from a text description using qianwen-video-generation?
qianwen-video-generation converts text prompts into videos through its primary text-to-video capability. Submit your text description as a request, and the skill routes it to the appropriate Wan model. Since all operations run asynchronously, you'll receive a request ID and must poll for completion. The skill includes detailed prompt engineering guides to help you craft descriptions that produce the best results.
Can qianwen-video-generation animate a static image into a moving video?
Yes. qianwen-video-generation supports converting static images into animated videos. You can upload a single image, and the skill will generate motion and animation around it. This is one of several input modes the skill auto-detects; it also handles frame transitions, text prompts, and video editing workflows all within the same platform.
What does qianwen-video-generation do with first frame and last frame video transitions?
qianwen-video-generation can create smooth transitions and animations between two images by treating them as first and last frames. The skill synthesizes the motion and content flow between these reference points. This capability is part of its multi-input approach, which also includes text-to-video, image animation, character role-play, and video repainting workflows.
Does qianwen-video-generation support video editing and recomposition with AI?
Yes. qianwen-video-generation includes video editing and repainting capabilities, allowing you to recompose and edit existing video content using AI. The skill auto-detects your task type and routes requests to the right model. All operations are asynchronous; submit your request and poll for completion using the reference guides provided for polling patterns and media workflows.
Can I generate character-driven videos with role-play scenarios in qianwen-video-generation?
qianwen-video-generation supports character-driven video generation with role-play scenarios. This is one of five primary input modes the skill handles, alongside text prompts, image animation, frame transitions, and video editing. The skill's auto-detection routes your request to the appropriate Wan model, and detailed guides cover prompt engineering for character-focused content.
What license does qianwen-video-generation use?
qianwen-video-generation is released under the Apache-2.0 license, allowing broad use, modification, and distribution under the terms of that open-source license.
SKILL.md
rendered from the published skill — quoted content, verbatim
> Agent setup: If your agent doesn't auto-load skills (e.g. Claude Code), > see agent-compatibility.md once per session.
Qwen Video Generation
Generate videos using Wan models. All tasks are asynchronous — submit, then poll until completion. This skill is part of QianWen-AI/qianwen-ai.
> ⚠️ Critical Parameter Differences by Mode: > - kf2v (First+Last Frame): Duration is fixed at 5 seconds — other values will fail. Output is
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 15 files
skills/video/qianwen-video-generation/SKILL.md
skills/video/qianwen-video-generation/references/agent-compatibility.md
skills/video/qianwen-video-generation/references/api-guide.md
skills/video/qianwen-video-generation/references/examples.md
skills/video/qianwen-video-generation/references/execution-guide.md
skills/video/qianwen-video-generation/references/merge-media.md
skills/video/qianwen-video-generation/references/polling-guide.md
skills/video/qianwen-video-generation/references/prompt-guide.md
skills/video/qianwen-video-generation/references/request-fields.md
skills/video/qianwen-video-generation/references/sources.md
skills/video/qianwen-video-generation/references/workflows.md
skills/video/qianwen-video-generation/scripts/gossamer.py
skills/video/qianwen-video-generation/scripts/qianwen_lib.py
skills/video/qianwen-video-generation/scripts/video.py
skills/video/qianwen-video-generation/scripts/video_lib.py