Please note that there are current limitations when using multiple trained models in a single image:
Current Capability
Using 2 or more different trained models of people or products in one image is not yet reliable
While generation may work, the quality of results can be inconsistent
What This Means
For best results, use one trained model per image
This applies specifically to combining different trained models (e.g., two different people or two different products)
This does not affect generating multiple instances of the same model
Future Improvements
Our team is actively working on improving this functionality
Stay tuned for updates that will enable reliable multi-model generation
For now, we recommend focusing on single-model generations to ensure the highest quality results.