ImageNet 512

'ImageNet 512' is a benchmark dataset used to evaluate the performance of image generation models, containing images resized to 512x512 pixels.

Papers using ImageNet 512 (3)

Unified Latents (UL): How to train your latents2026PixelDiT: Pixel Diffusion Transformers for Image Generation2025There is No VAE: End-to-End Pixel-Space Generative Modeling via Self-Supervised Pre-training2025