Choose Your AI Generation Engine

Flat-to-Model turns your flat-lay product photos into photorealistic on-model imagery, and different AI engines interpret the same inputs in different ways. Some lean clean and editorial, others lean dramatic and creative. Until now we picked one for you. Starting today, you can pick the one whose aesthetic fits your brand, or stay on Auto and let us choose.

This rolls out across both products. Flat-to-Model offers Nano Banana Pro and Seedream. Model-Swap offers Onda and Nano Banana Pro. Flat-to-Model is where the aesthetic difference between engines is most visible, so that's what this post focuses on. A short section at the end covers the same dropdown in Model-Swap.

What's new

A new Generation model selector appears in the Flat-to-Model review step, right alongside the existing pose, background, and aspect ratio controls. The same choice is available via the API as an optional model field on the request's options object. If you don't set it, the default is Auto. That's exactly the behavior every job had before.

The Flat-to-Model engines

Three options:

• Auto (default): runs Nano Banana Pro first. If the input trips the safety filter, we transparently reroute to Seedream so the job still completes.
• Nano Banana Pro: Google's latest image model. Strong garment fidelity and clean photorealistic skin. No fallback on refusal.
• Seedream: ByteDance's image model. Often more dramatic lighting and a more creative interpretation of the styling. No fallback on refusal.

Same inputs, different engines

Here's the same flat-lay trio, the same identity, and the same pose, rendered by each engine.

Same garments, same identity, same pose: seated on a wooden stool against a warm ecru plaster wall. Nano Banana Pro renders the scene with a fully frontal composition, crisp collar lines, and cooler lighting that holds the icy blue of the shirt. Seedream nudges the model into a three-quarter turn, softens the tones, and pushes the backdrop a touch warmer, with a slightly more relaxed drape on the shirt. Neither is better. They're different aesthetic defaults, and picking the one that fits your brand catalogue is the point.

Which engine produced each output

Once a job finishes, you can see which engine was actually used for every result.

In the app, a small engine icon sits next to the processing time on each output card. Hover over it for the engine name.

In the API, each result in GET /jobs/<id>/results carries a model_used field:

{
  "job_id": "8c9f1a...",
  "status": "completed",
  "results": [
    {
      "image_index": 0,
      "group_index": 0,
      "model_used": "nano_banana_pro",
      "output": { "full_size": "https://..." }
    },
    {
      "image_index": 1,
      "group_index": 0,
      "model_used": "seedream",
      "output": { "full_size": "https://..." }
    }
  ]
}

This matters most when you submit a job on Auto. If you asked for four outputs and three come back as nano_banana_pro while one comes back as seedream, you instantly know that single slot hit the safety fallback. No guessing, no support ticket.

Using it via the API

Pass model inside the options object on your Flat-to-Model request:

{
  "project_id": "...",
  "images": ["uuid1", "uuid2", "uuid3"],
  "identity_code": "default-pro-...",
  "instructions": [ /* ... */ ],
  "options": {
    "model": "nano_banana_pro"
  }
}

Valid values: "auto" (default), "nano_banana_pro", or "seedream".

In our public Python integration, it's a single CLI flag:

python flat_to_model.py \
  --input-folder SKU/ \
  --identity-code <your-identity-code> \
  --model nano_banana_pro

Full reference is in the Flat-to-Model API docs.

Model-Swap gets a dropdown too

The same Generation model selector is now in the Model-Swap review step.

The default is Onda, PiktID's proprietary swap engine, which is what every Model-Swap job has run on to date. The alternative is Nano Banana Pro, available for teams who want to try a different aesthetic on their swaps. model_used works exactly the same way: every result in GET /jobs/<id>/results tells you which engine produced it.

Pass swap_options.model on a Model-Swap job, or --model onda | nano_banana_pro on the Python integration.

"Different engines are different tools. For a long time we picked one for you, and that was the right default. But brands and agencies who run hundreds of SKUs a week know their aesthetic, and they deserve the choice. Exposing engine selection turns On-Model from a black box into a controllable production pipeline, and model_used on every output makes the choice auditable."

— Nunzio Alexandro Letizia, Co-founder at PiktID and creator of On-Model

Try it now

Open any Flat-to-Model or Model-Swap job and look for the Generation model dropdown in the review step. Pick an engine, generate, and compare the result against Auto.

New to On-Model? Start with the Flat-to-Model guide or explore presets for even more control over the output.