AI Paper Atlas

Research map / Benchmarks / ImageNet 512×512

ImageNet 512×512 leaderboard

15 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ImageNet 512×512. Best FID: 1.45 (NiT-XL (197B tokens), NeurIPS 2025).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

ImageNet 512×512 · FID

15 papers, each with its best reported number for its own method; lower is better.

#FID ↓MethodPaperTopic
11.45NiT-XL (197B tokens)
DiT-XL/2
Native-Resolution Image Synthesis
NeurIPS 2025 · Table 1
Autoregressive and latent generation
21.45EDM2-XXL + Interval CFG + REG
EDM2-XXL
REG: Rectified Gradient Guidance for Conditional Diffusion Models
ICML 2025 · Table: Class-conditional Image
Fast diffusion sampling
31.62TexTok-256 + DiT-XL
DiT-XL
Language-Guided Image Tokenization for Generation
CVPR 2025 · Table 2(b)
Autoregressive and latent generation
41.77eMIGM-LEffective and Efficient Masked Image Generation Models
ICML 2025 · Table: Image generation result
Autoregressive and latent generation
51.8PaGoDA
EDM
PaGoDA: Progressive Growing of a One-Step Generator from a Low-Resolution Diffusion Teacher
NeurIPS 2024 · Table: Experimental results of
Fast diffusion sampling
61.83CoSIM-L, 4 NFE
EDM2-L-FID
Continuous Semi-Implicit Models
ICML 2025 · Table: Conditional generation
Fast diffusion sampling
72.63VAR-d36-sVisual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
NeurIPS 2024 · Table (ImageNet 512×512 compar
Autoregressive and latent generation
82.86DiT-XL/2 + iSSD, 50 steps
DiT-XL/2
Diffusion Model is Effectively Its Own Teacher
CVPR 2025 · Table 2
Fast diffusion sampling
92.99HMAR-d24HMAR: Efficient Hierarchical Masked Auto-Regressive Image Generation
CVPR 2025 · Table 2
Autoregressive and latent generation
103.61DyDiT-XL λ=0.7
DiT-XL 512×512
Dynamic Diffusion Transformer
ICLR 2025 · Table: Image generation at 512
Fast diffusion sampling
5 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.