ColabzColabz
Open app
Welcome
  • Welcome to Colabz
  • Sign-up walkthrough
  • Tour the dashboard
AI Photoshoot
  • Make your first image
  • The Photoshoot Wizard
  • Prompt patterns
  • Reusing products and faces
  • Quality tiers (1K, 2K, 4K)
Video Studio
  • Make your first video
  • Animate a still
  • Picking a vibe
Photo Magic
  • Photo Magic overview
  • Remove Background
  • Replace Background
  • Extend
  • Erase
  • Upscale a photo
  • Model Swap
  • Advanced Edit
Brand Library
  • Products explained
  • Adding a product
  • Faces explained
  • Generate a face
  • Styles explained
  • Trained models
Studios
  • Studios explained
  • Picking a Studio
  • Custom Studios coming soon
Workspace
  • Your library
  • Creating boards
  • Saving and sharing
  • Inviting members
  • Roles explained
Plans & Billing
  • Plans overview
  • How credits work
  • Top up credits
  • Subscription and billing
  • Launch promo (April 2026)
Free Shot
  • The free Shot tool
Help
  • Help center
  • Credit questions
  • Getting help
  • Rules and policies
Brand Library

Trained models

Trained models are LoRAs the AI latches onto. They produce more consistent results than reference photos alone. 150 credits, takes 5 to 15 minutes.

Train one#

  1. Gather your reference set. 5 to 20 photos for products, 10 to 30 for styles, 10 to 20 for faces.

  2. Open /app/train and click Create New Model.

  3. Pick the type. Product, Style, or Face.

  4. Upload the references. Drag and drop. The trainer validates count and format.

  5. Name and start training. 150 credits. Comes back in 5 to 15 minutes.

  6. Use it in any prompt. @your-handle and combine with Studios.

Status meanings#

  • Training. The model is being prepared.
  • Ready. Usable via @ in prompts.
  • Failed. Training did not converge. Credits refunded automatically.

Tip: Pick memorable handles. You will type these dozens of times.

My trained model drifts off-product#

Mention the model twice in long prompts (early and late). If it still drifts, retrain with a tighter, cleaner reference set.

Training failed#

Usually a reference set problem: too few images, mixed subjects, harsh lighting. Credits are refunded automatically. Clean up the set and retrain.

Related actions

  • Products explained
  • Styles explained
  • Faces explained
Previous
Styles explained
Next
Studios explained
On this page
  • Train one
  • Status meanings
  • My trained model drifts off-product
  • Training failed