RaspizdAI/ToiletGAN1

ToiletGAN V1 (128x128)

Model Overview

• Architecture: DCGAN (Deep Convolutional Generative Adversarial Network)
• Resolution: 128x128 pixels (RGB)
• Latent Dimension: 100
• Training Steps: 4,500 global steps

Description

ToiletGAN V1 is a lightweight generative model built under the RaspizdAI initiative.

🔗 Try it live in your browser: ai.amogus.bond/toiletGAN/v1

Available Files & Variants

• toiletV1_128-IR8.onnx — Legacy-compatible version with IR version 8, optimized for older client runtimes.
• toiletV1_128.onnx — Standard modern export version.

Upcoming

• ToiletGAN V2 — Coming soon with improved architecture and higher visual fidelity!

License & Attribution

This model is licensed under CC-BY-SA-4.0.

• You are free to share and adapt the material.
• You must give appropriate credit to RaspizdAI, provide a link to the license, and indicate if changes were made.
• You cannot remove the copyright/attribution or close the source of derivative models.