Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Text-to-Image Generation research benchmarks – Page 33 · SOTA2 Research

Text-to-Image Generation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

100 prompts (test)

0.282HPSv2

4

Feb 26, 2026

Rewritten prompts

0.8Bias

4

Feb 26, 2026

Context prompts

0.815Bias

4

Feb 26, 2026

0.797Bias

4

Feb 26, 2026

Occupation prompts

0.746Bias

4

Feb 26, 2026

Poster&SceneBench 1.0 (test)

93.2Poster Score

4

Feb 26, 2026

98.7L1 Score

4

Feb 26, 2026

1.61Rew

4

Feb 26, 2026

1.118Rew

4

Feb 26, 2026

Unspecified Dataset (test)

Vision-Enhanced LLM

12.5Training Time (hrs)

4

Feb 26, 2026

6.046Aesthetic Score

4

Feb 26, 2026

400 prompts (test)

29.053HPSv2

4

Feb 26, 2026

MeiGen-Image-FaQ

88Win Rate

4

Feb 26, 2026

31M image dataset (test)

5.9FID

4

Feb 26, 2026

Dalle3 HQC 1M+ (5,000 pairs)

49.9FID

4

Feb 26, 2026

3D spatial scratchpad

482D Spatial Score

4

Feb 26, 2026

ImageNet 256×256 (test val)

2.453FID

4

Feb 26, 2026

Manually constructed text prompt set 1.0 (160 prompts) (test)

31.99CLIP Prompt Score

4

Feb 26, 2026

Text-to-image Prompt: 'A creative bridge design'

2.47Vendi Score

4

Feb 26, 2026

0.541Text Alignment

4

Feb 26, 2026

Text-text similarity attribute binding suite

Self-Coherence Guidance

85.2Coarse-grained Binding Score

4

Feb 26, 2026

Attribute Binding Style

79.9BLIP-VQA Score

4

Feb 26, 2026

Attribute Binding Fine-grained

62.3BLIP-VQA Score

4

Feb 26, 2026

Attribute Binding Coarse-grained

67.9BLIP-VQA Score

4

Feb 26, 2026

GenEval 256x256

DiffMoE-E16-T2I-Flow (w SFT)

0.96Single Object Composition

4

Feb 26, 2026

Page 33 of 39

Popular datasets

GenEval MS-COCO DrawBench COCO T2I-CompBench

Follow for updates

Join SOTA2 Contact