Multi-scene video generation(Bouncy Walk)

This is a workflow for generating a video with continuous scenes based on a set plot. It can keep the character, lighting, and color tone consistent across scenes (though you can also set different backgrounds, you'll just need to use the right prompts). The workflow also includes the System Instructions I used to generate the storyboard scene prompts and video prompts, which can help you quickly generate your own using an LLM. For important details, please see the testing and usage instructions below.

💻I've already set up the complete ➡️ workflow for you. This will let you quickly try it out online and avoid the hassle of insufficient VRAM or local plugin conflicts.

🎁Bonus: If you're signing up for the first time, you can get 1,000 free RH Coins by using my link and the invite code ➡️rh-v1182. Plus, you'll get another 100 RH Coins for logging in daily.

Workflow Testing and Usage Instructions:

1. First, a big thank you to ai_build_art for releasing the "Bouncy Walk" Lora. You can download their Lora at this link.

2. The Qwen-image-edit_2509 model is already capable of generating images with different scenes on its own. The reason I'm using the "next-scene_lora_v1-3000" Lora in this workflow is that it's better at generating multi-scene storyboard images with different camera angles or backgrounds.

3. I've attached two sets of System Instructions to the workflow. You can use them with an LLM you're familiar with to quickly generate your own storyboard scene and video prompts. I recommend using Gemini 2.5 Pro to generate these prompts.

4. I have already marked the download links for the main models used in the workflow, so you can easily download them for local use. If you're using this workflow online, just load the corresponding images and run the workflow. For other detailed instructions, please see the notes inside the workflow.