Skip to main content
All CollectionsImage Generation
Product shots with Custom AIs
Product shots with Custom AIs

Custom model training in Kive lets you train custom AI models using your images. Generate high-quality, on-brand product photography using either Standard or Advanced mode.

Olof Lindh avatar
Written by Olof Lindh
Updated over a month ago

What is Custom AI?

Custom AI, or model training, teaches Kive's AI to recognize and reproduce specific products, objects, people, and styles. Once trained, these models can be used in Product Shots to create consistent and on-brand imagery.

Product Shots is available on the Pro plan, and Style Models are available on Enterprise plans.

Training a Custom AI model

You can choose between two modes for training object models: Standard Mode and Advanced Mode.

Standard Mode

  • Quick and simple: Requires as little as one clear image.

  • Generates good results for the exact angle uploaded.

  • Uses fewer AI credits and has faster processing times.

Advanced Mode

  • Superior quality for soft objects (e.g., clothing) and more creative angles.

  • Requires multiple high-quality images from varied angles.

  • Takes longer to process and consumes additional AI credits.

Tips for Training (Advanced Mode)

Selecting images

Use at least 5 images per subject (10+ for best results). Clear, well-lit images help the AI learn accurately.

  1. Distinct backgrounds

    Ensure the object contrasts clearly with the background. Avoid matching background colors.

  2. Diverse images

    Different backgrounds and lighting conditions teach the AI to recognise your product in varied settings.

  3. Multiple angles

    Capture essential perspectives—front, side, high, low—relevant to the images you'll generate. You will be able to lock generation to the angles you select.

  4. Varied lighting

    Include natural, artificial, and other lighting scenarios for adaptable models.

  5. High quality

    Sharp images with at least 1024x1024px resolution. Avoid blurry or pixelated images.

  6. Close-ups

    Include close-up shots of logos, labels, and critical details for precision.

Example input images

Using a Custom AI model

After training, use your Custom AI by tagging it in prompts with @. For example: "Photo of @Acme Serum".

Examples

photo of @Acme Serumsurrounded by flowers, flowers everywhere, with lush rain forest background, deep in a jungle, lush nature, cinematic scene, ultrasharp photo, cinematic, Cinestill 800T, shot by Roger Deakins, shot on Kodak Portra, beautiful light, dawn

photo of @Acme Serum surrounded by flowers, flowers everywhere, with lush rain forest background, deep in a jungle, lush nature, cinematic scene, ultrasharp photo, cinematic, Cinestill 800T, shot by Roger Deakins, shot on Kodak Portra, beautiful light, dawn

photo of @Acme Serumstanding on a shiny graphite crystal crest, peak of a graphite mountain, dark dusk, blue hour, low light, cold light, shot on ultrasharp 35mm film, shot with Leica, diffracting light, glow, black mist diffusion filter

photo of @Acme Serum standing on a shiny graphite crystal crest, peak of a graphite mountain, dark dusk, blue hour, low light, cold light, shot on ultrasharp 35mm film, shot with Leica, diffracting light, glow, black mist diffusion filter

Did this answer your question?