HyperAI

Main

GPU

Console
Studio
Docs
Pricing

Pulse

News

Resources

Papers
Notebooks
Datasets
Wiki

Benchmarks

SOTA
LLM Models
GPU Leaderboard

Community

Conferences

Utility

About Terms of Service Privacy Policy
English

Command Palette

Search for a command to run...

HyperAI
SOTA
Image Generation

Image Generation

Creating realistic or artistic images from scratch, often conditioned on text descriptions or other modalities.

Leaderboard

100 models total

Benchmarks

EDM-G++ (conditional)

ImageNet 256×256

ImageNet 256x256

GTAV-to-Cityscapes Labels

ImageNet 512x512

ImageNet 128x128

Transition Matrix

ImageNet-256×256

LSUN Bedroom 256 x 256

LSUN Churches 256 x 256

ImageNet 512×512

ImageNet 64×64

DreamBooth-dataset

FFHQ 1024 x 1024

Polarity-StyleGAN3

CelebA-HQ 256x256

Vision-aided GAN

ImageNet-512×512

custom facial personalization dataset (50 identities)

MS-COCO 2017 validation set

COCO-2017 validation set

Attend-and-Excite Benchmarks

CelebA-HQ 1024x1024

COCO validation set

Fundus Fluorescein Angiogram Photographs & Colour Fundus Images of Diabetic Patients

COCO Captions 2014

MS-COCO 2014 validation

Vision-aided GAN

StyleGAN2 + ADA

CelebA-HQ 128x128

LSUN Cat 256 x 256

Anime character pose dataset

Dayton (256×256) - aerial-to-ground

DreamBooth benchmark

Pick-a-Pic test set

Visual Genome 64x64

Wiki-Art dataset

AnyText-benchmark

COCO-Stuff 128x128

COCO-Stuff 256x256

COCO-Stuff 64x64

Custom Concept 101

Dayton (64×64) - aerial-to-ground

Dayton (64x64) - ground-to-aerial

MS-COCO 2017 validation split

Stable-Diffusion-Prompt

StyleDrop style set

Visual Genome 128x128

LayoutDiffusion

VoxCeleb2 - 1-shot learning

Fast Bi-layer Avatars (medium size)

Aesthetic-Eval@4096

Vision-aided GAN

AI-generated Images

Dayton (256×256) - ground-to-aerial

DreamFusion gallery

Edge-to-Handbags

ImageNet (512×512)

LAION-Aesthetic-6+

LayoutBench-COCO - Number

LayoutBench-COCO - Combination

LayoutBench-COCO - Position

LayoutBench-COCO - Size

LSUN-Bedroom 256×256

LSUN-Church 256×256

LSUN Horse 256 x 256

MovieLens Latest Small

MS-COCO 2014 validation set

Multilingual-Cultural (MC)

Pokemon 256x256

Poster-Benchmark

CyCADA pixel+feat

Textual Inversion dataset

Visual Genome 256x256

LayoutDiffusion

30K object-centric images

5k evaluation dataset

A3D object pairs

CelebA-HQ 64x64

CIFAR-10 (10% data)

DiffAugment-StyleGAN2

CIFAR-10 (20% data)

DiffAugment-StyleGAN2

COCO validation dataset

COCO-2014 validation set

COCO-2017-val 30K prompts

COCO'14 validation

Creative Architecture

DDPO animal prompts

DiffDoctor treating benchmark

DIODE-Outdoor (256 × 256)

DressCode-Upper

EcoMapper dataset

Edges→Handbags (64 × 64)

Freiburg Forest Dataset

Human and Hand Generation (H&H) Dataset

ImageNet-64×64

internal human image dataset

LSUN Bedroom 64 x 64

WGAN-GP + TTUR + Alex-Adam

PixArt-XL-2-256-MS

Pokemon 1024x1024

Shop2Model (VITON-HD)

StyleGAN2 with DiffAugment

Stable Diffusion 3

Stable Diffusion 3.5

stereotype prompts

test subset (all brands used for training)

Text-to-Image Generation

Unseen brands test set

512×2048 panorama

Aesthetic Score

AFHQ-D 256×256

ArchiMetricsNet

BDD100K-Subsets

CelebA-HQ + FFHQ

CelebAHQ 256×256

ChestXray14 1024x1024

Class-Conditional ImageNet 256×256

COCO IFG benchmark

Complex-Benchmark

Conceptual Captions

Conditional ImageNet 128×128

Creative Creatures

Custom 680-image dataset

Custom benchmark

DreamFusion prompts

EMU-Edit test set

Existence and Attribute Binding (A&E) Dataset

FFHQ (256×256)

FFHQ (64 × 64)

FFHQ 1024×1024

Google Scanned Objects (GSO)

Home Action Genome

HPDv2 photo+painting prompts

ImageNet (64 × 64)

ImageNet (cond.)

ImageNet 128×128

StyleGAN2 + NoisyTwins

IMM Face Dataset

iNaturalist-2019

Inert-CompBench

InnoComposer-Bench

InstructPix2Pix

internal dataset

LAION-Aesthetics

LAION-Aesthetics V2

LAION-Aesthetics V2 6.5+

LIGER benchmark

LSUN Bedroom 128 x 128

LSUN-Bedroom-256

LSUN-Bedrooms 256×256

LSUN Car 512 x 384

LSUN-Churches 256×256

MapillaryVistas 2048×2048

Mix-of-Show dataset

MMSVG-Illustration

MS COCO 512x512

MS COCO Dataset

MS-Bench-Random

MS-COCO 2017 captions

MS-COCO 30k validation set

MSCOCO-2014 30K

multi-subject generation

Occupational Prompts

OpenImages-Test

Oxford 102 Flowers 256 x 256

Pic-a-Pick prompts

REQUITE-Prostate

single-subject generation

Split-miniImageNet

standardized image prompt set

StyledTextSynth

SubjectSpatial200K

synthetic human dataset

T2ICompBench-3D

Text-to-3D Generation

text-to-3D sketch prompts

VoxCeleb1 - 1-shot learning

Few-shot Adversarial Model

VoxCeleb1 - 32-shot learning

Few-shot Adversarial Model

VoxCeleb1 - 8-shot learning

Few-shot Adversarial Model

VoxCeleb2 - 8-shot learning

Zero-shot LAION-5k 1024×1024

100 Artistic Styles

100-shot Grumpy

10K High-Quality Test Set

15 prompt pairs

20-concept dataset

200 prompt and bounding-box pairs

25% ImageNet 128x128

310k synthetic images and caption pairs

32 diverse 3D objects

40 text prompts

AI-Generated Prompts

Animation Dataset

Artist-created Mesh

Artist-created mesh benchmark

Attend-n-Excite prompts

Bitmoji dataset

Blender-rendered synthetic dataset

blurred VOC 2012

Bosonplus-night

CelebA (64 × 64)

CelebA_Large_5X

CelebA-HQ 256×256

CelebA-HQ 512x512

Challenging viewpoints

CHAOS liver (CT)

Chinese benchmark

Chinese cultural dataset

Chinese SVG dataset

ChineseDrawText, DrawTextCreative, TMDBEval500

CIFAR-10 (Unconditional)

CIFAR-10 (Conditional)

Cityscapes-25K 256x512

Cityscapes-5K 256x512

COCO validation

COCO2014 5K validation

COCO2017 validation

collected dataset of 25 identities

Colorized MNIST

ConsiStory-Human

Controllable Generation

Custom dataset (10 prompts × 5 variations × 6 objects)

Custom dataset of 12 scenarios

Custom Hand Test Set

Custom image generation evaluation dataset

Custom Indoor Scenes

Custom Test Set

Custom test set (800 meshes)

Custom101-StyleDrop

Decathlon heart (MRI)

Decathlon liver (CT)

DeepFashion In-Shop Clothes Retrieval Benchmark

design websites

DreamBench Plus

DreamBooth dataset and Custom-Concept101

Dreambooth-StyleDrop

DreamFusion & MVDream prompts

DreamFusion gallery prompts

DressCode dresses

DressCode lower-body

DressCode upper-body

Edges2handbags (64 × 64)

Spiking-Diffusion

emotional prompts dataset

English benchmark

English user inputs

evaluation data set

Face-blurred ImageNet-256

FastComposer benchmark

FFHQ (1024×1024)

FFHQ & CelebA-HQ

FFHQ 64 × 64 (Unconditional)

FFHQ 512 x 512 - 4x upscaling

FFHQ 64x64 - 4x upscaling

FFHQ-Sunglasses

General objects

Generative Mesh

GenText-Microstruct

Geometries 2000

GlyphMM-benchmark

Google Scanned Object

GPT-4 generated prompts

Great Lakes Invasives Network (GLIN) fish dataset

GSO T2T benchmark

Hand-Drawn Dataset

HPS v2 benchmark

Human customization

IAM Offline Handwriting Database

image-to-3D sketch

image-to-texture

ImageNet (256×256) validation subset

ImageNet (32 × 32)

ImageNet 1024×1024

ImageNet 256 × 256 (Conditional)

ImageNet 256 256

ImageNet 64 × 64

ImageNet 64 × 64 Cond.

ImageNet-1K classes

ImageNet 256x256 - 1% labeled data

ImageNet 256x256 - 1 labeled data per class

ImageNet 256x256 - 2 labeled data per class

ImageNet 256x256 - 5 labeled data per class

ImageNet-R-TI2I

Imbalanced CIFAR10

Indian Celebs 256 x 256

Internal Benchmark

LAION subset (10k image-text pairs)

LAION-Aesthetics 6.5+

LAION-Aesthetics-6+

LAION-Aestheticsv2 6.5plus

LAION-High-Resolution

Landmark photo sets (6 landmarks)

Landscapes 256 x 256

Lay2Story-Bench

LPI simulation dataset

LSUN Bedroom 256×256

LSUN Bedrooms 256×256

LSUN Church (256×256)

LSUN Church (64 × 64)

LSUN Car 256 x 256

LSUN tower 64x64

MagicBrush test set

Male-to-Female (128x128)

Male-to-Female (64x64)

MatSynth + Deschaintre et al.

MiniText-Benchmark

MIT Indoor Scenes

MS-COCO 256×256

MS-COCO validation set

MS-COCO_Large_5X

Multi-Entity Draw Bench

Multi-Human Benchmark

multi-subject driven generation benchmark

Multi-Subject Personalization

Natural Scenes Dataset (NSD)

noisy-ADE20K-DS

noisy-ADE20K-Edge

noisy-ADE20K-Random

novel human-object dataset

Objaverse and game assets

Objaverse subset

Objaverse T2T Benchmark

objaverse, shapenet, Stanford 3D Scanning Repository

occupation prompts

Open Images validation set

ORIBENCH-Single

Oxford 102 Flowers 128x128

PanFusion prompts

PanoLlama benchmark dataset

Pre-set Collection

Proprietary Dataset

Reference dataset

ROVI validation set

Satellite-Buildings 256 x 256

Satellite-Landscapes 256 x 256

SceneVTG-benchmark

SD-generated images

Simulator video dataset

Something-Something (V2)

Stable Diffusion Prompts Dataset

Stable Diffusion v1.5

Stable-Diffusion-Prompts

StyleAligned benchmark

StyleAligned-DreamBooth

Subject-Insertion

Subjectplop-StyleDrop

SyncTweedies mesh and prompt pairs

Synthetic dataset

T2I-CompBench++

T2I-CompBench++ (3D Spatial)

T3bench+GPTEval3D

TexTalker4D-Test-A

US cities urban layout dataset

Vector-Fusion prompts

VectorFont (Test)

Visual Capitalist

VITON-HD In-the-Wild

VoxCeleb2 - 32-shot learning

VRoid-Hub garment dataset

web-sourced test set

WikiArt and StyleDrop

Wild-to-Cat (64x64)

Build the Future of Artificial Intelligence

About

About Us Blog Support Dataset Help

Products

News Papers Notebooks Datasets Wiki

Links

© HyperAI

GitHub Discord X (formerly Twitter)

HyperAI AgentBeta

HyperAI Agent is in beta. Answers may be inaccurate, so check important details.

Conversations are deleted 14 days after they start.

What can I help with?

Find resources, learn the platform, or manage your workspaces