Suno v6 + Seedance 2.5 is NUTS at Making AI Music Videos

What this lesson covers

Dan Kieft demonstrates an end-to-end workflow for producing AI music videos, starting with audio generation and moving into visual asset creation. He outlines a three-step character development pipeline combining LLM prompting, image reference sourcing, and multi-model generation.

Key takeaways from the creator

AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.

1. 02:20 ↗

   Start music production by creating an instrumental beat first before uploading it as an audio reference to remix genre, tempo, and lyrics in Suno

2. 05:12 ↗

   Structure character design into a three-step workflow focusing on the face, outfit styling, and character sheets to maximize visual control

3. 05:34 ↗

   Define actor traits including physical appearance, makeup, and personality backstory before prompting to maintain consistent persona

4. 06:24 ↗

   Utilize an LLM-based director prompt to convert unstructured character ideas into structured image prompts

5. 07:58 ↗

   Source external visual references for outfits and use an LLM to reverse-engineer prompts matching the reference aesthetic

Workflow outlined in the video

1. Compose an instrumental track in Suno to establish the core rhythm and beat structure
2. Upload the instrumental reference into Suno to remix the genre and apply custom lyrics
3. Input character concepts into an LLM director template to refine visual and personality specifications into prompts
4. Generate face variations across available image models in Higgsfield to select the optimal character look
5. Upload reference outfit images to an LLM to generate detailed clothing prompts for character styling

Before you use this workflow

These notes describe the source video at its publication date. Model access, pricing, connectors and interfaces may have changed. Check the original source and the provider’s current documentation before installing an add-on, connecting an account or spending credits.

ReelStack has not independently tested this workflow. Preview one representative shot and check motion, continuity and output quality before applying it to a full production. No result, cost saving or model capability is guaranteed.

Explore the tools

How to interpret the numbers

42,646 views captured 6 October 2026. ReelStack recommendation score: 62/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.

Read our methodology and limitations →