Notes by ReelStack · AI-assistedUpdated 20 September 2026

I Made an iPhone 18 Ad That Got 1M Views in 2 Days (Tutorial)

Offers AI filmmakers a clear, cost-conscious multi-model pre-production framework for locking down character and product consistency before generating video sequences.

Video thumbnail: I Made an iPhone 18 Ad That Got 1M Views in 2 Days (Tutorial)
Original YouTube video

kaye.creatives

Published

Watch the original video ↗

What this lesson covers

Kaye breaks down her workflow for creating a viral speculative iPhone 18 commercial using generative AI. She explains how to craft 360-degree character and product model sheets with Claude and generate consistent assets using specialized models in Higgsfield Cinema Studio.

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. 01:20 ↗

    A character sheet acts as a 360-degree reference blueprint locking height, facial features, and styling across angles to maintain visual consistency across generations.

  2. 02:14 ↗

    Feeding a personal selfie and an outfit reference into Claude allows you to automatically generate three-panel (front, profile, full-body) character sheet prompts without manual writing.

  3. 03:03 ↗

    Soul 2.0 is optimal for generating human assets with realistic skin texture without synthetic artifacts while keeping generation credit costs minimal.

  4. 06:40 ↗

    Cream 5.0 Pro provides sharp graphic precision suitable for rendering consistent product model sheets from multiple angles.

  5. 08:19 ↗

    Soul Cinema is tailored for location generation with architectural lighting, realistic optics, and natural depth.

Workflow outlined in the video

  1. Find outfit and style references on visual platforms like Pinterest.
  2. Upload reference images and selfies into Claude to generate structured three-panel character sheet prompts.
  3. Input the generated prompts into Higgsfield Cinema Studio using the Soul 2.0 model in a 16:9 ratio in batches of four to select the best character likeness.
  4. Generate product model sheets by feeding leaked or concept reference images into Claude and rendering the prompt via Cream 5.0 Pro for clean industrial design details.
  5. Create cinematic environment plates using Soul Cinema to establish architectural lighting and camera depth before moving into video generation.

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

7,340 views captured 20 September 2026. ReelStack recommendation score: 40/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.

Show the calculation
  • 65% performance: 0.11× current views vs the channel's recent median; same-age history is not yet available. View-evidence factor 88% (views / (views + 1,000))
  • 25% freshness: 96/100 with a 14-day half-life
  • Momentum pending: collecting daily snapshots; its 20% weight goes to observed performance, not free points
  • 10% engagement: 100/100; likes + 4× comments, smoothed with a 500-view neutral prior

Read our methodology and limitations →

Powered by ReelStack

Help keep this running

Your tip funds servers, models, and the time it takes to ship new tools faster. Set any amount below — every bit helps.