How I Built a Car Commercial With AI (Every Prompt I Used)
This video gives commercial creators a direct blueprint for combining LLM prompt generation with multi-model image workflows to maintain strict visual brand consistency.
What this lesson covers
Adio demonstrates a complete, three-step production workflow for building an AI car commercial using Higgsfield, Claude, and specialized video generation models. He provides detailed techniques for tagging project assets, comparing model outputs for realistic character swaps, and enforcing uniform visual color palettes.
Key takeaways from the creator
AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.
- 02:14 ↗
Generating hero product shots on a plain gray backdrop simplifies subsequent image compositing and background blending.
- 02:47 ↗
Tagging assets with element handles in Higgsfield and linking identical names in Claude allows for automated prompt building and element attachment.
- 03:48 ↗
Running identical prompts across specialized models like SeaArt Dream Pro 5.0 and Niji Journey Pro reveals distinct strengths in fabric rendering versus face accuracy.
- 05:00 ↗
Generating expressive character poses in creative models like Soul Cinema and then performing face swaps yields more natural cinematic performances.
- 09:03 ↗
Uploading a reference image into the color transfer feature of Higgsfield Cinema Studio locks in a unified lighting and color palette across all commercial shots.
Workflow outlined in the video
- Shoot or generate product assets on a neutral gray backdrop to make background swapping easier.
- Create dedicated subfolders for each scene asset iteration inside Cinema Studio to maintain asset organization.
- Assign '@' element handles to reference images in Higgsfield and reference those same handles when prompting Claude.
- Perform initial character action renders using Soul Cinema, then apply face swapping in Niji Journey Pro or SeaArt Dream Pro to perfect character realism.
- Cut out faces from full-body renders using a photo editor to fix facial variations without re-generating full sheets.
- Apply color transfer using a central reference image to standardize color grading across all generated video clips.
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
185,538 views captured 27 September 2026. ReelStack recommendation score: 47/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.
Show the calculation
- 45% performance: 1.65× views vs 6 other channel videos observed at a similar age. View-evidence factor 99% (views / (views + 1,000))
- 25% freshness: 7/100 with a 14-day half-life
- 20% momentum: 783 views/day vs 343 channel baseline (6 comparable uploads), with the same view-evidence factor
- 10% engagement: 35/100; likes + 4× comments, smoothed with a 500-view neutral prior