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 23 · SOTA2 Research

Text-to-Image Generation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

SD-3.5-Medium + PickScore

80.5Attribute Adherence Score

7

May 19, 2026

SD-3.5-Medium + AutoRule (PickScore)

85.5Attribute Adherence

7

May 19, 2026

FLUX.1 T=50 (dev)

64.61GenEval

7

May 19, 2026

87.32GenEval

7

May 19, 2026

0.316Non-spatial Fidelity

7

May 15, 2026

92Accuracy (2 objects)

7

May 15, 2026

Memorized Prompts

0.58SSCD

7

May 6, 2026

Pick-a-Pic Unique V1 (val)

21.48PickScore

7

Apr 29, 2026

StableDiffusion-3.5

34.97CLIPScore

7

Apr 21, 2026

75.5SR (Gemini)

7

Mar 31, 2026

Dissimilar Background Biases

92SR (Gemini)

7

Mar 31, 2026

Similar Textures

87.5Success Rate (Gemini)

7

Mar 31, 2026

Extended Many Objects

85.5Success Rate

7

Mar 31, 2026

78SR (gemini)

7

Mar 31, 2026

TreeOfLife 500 species 1M

29.87FID

7

Mar 30, 2026

4.57FID

7

Mar 27, 2026

ShareGPT-4o-Image SD3-Medium

35.185CLIP Score

7

Mar 26, 2026

54.72FID

7

Mar 18, 2026

User Study COCO-style benchmarks

8.075Aesthetic Quality (Aes)

7

Mar 17, 2026

GenEval and MS-COCO SDXL (test)

Hyper-SDXL + GAG

27.78CLIP Score (CS)

7

Mar 4, 2026

MS-COCO 30K 2017

22.15FID

7

Mar 4, 2026

SD template memorization 2.0

73.5SSCD

7

Mar 4, 2026

SD template memorization 1.4

0.776SSCD

7

Mar 4, 2026

SD 1.4 (verbatim memorization)

0.079SSCD

7

Mar 4, 2026

Qwen-Image-Lightning Evaluation Set

1.52Latency (s)

7

Mar 5, 2026

Page 23 of 39

Popular datasets

GenEval MS-COCO DrawBench COCO T2I-CompBench

Follow for updates

Join SOTA2 Contact