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-Video Generation research benchmarks – Page 8 · SOTA2 Research

Text-to-Video Generation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

NeuS-V 160-prompt human-evaluated subset 1.0

0.7PCC (Nature)

4

Feb 26, 2026

NeuS-V 360-prompt set 1.0 (full set)

0.716Nature Score

4

Feb 26, 2026

61.52FID

4

Feb 26, 2026

Unspecified Text-to-Video Dataset (test)

Text2Video-zero

0.732Clip-score

4

Feb 26, 2026

0.301Frame Consistency (F.C.)

4

Feb 26, 2026

Text-to-Video Generation prompts (test)

4.03QAlign Image Quality

4

Feb 26, 2026

ActivityNet (test)

86.6Action Faithfulness

4

Feb 26, 2026

User Study Publicly available samples from closed-source models

80Photorealism

4

Feb 26, 2026

WebVid-10M 2-million

48.3CLIP Score

4

Feb 26, 2026

WebVid-10M (val)

292.35FVD

4

Feb 26, 2026

82Object Class Fidelity

3

Jul 7, 2026

4.1SA Score

3

Jun 29, 2026

122.31Latency (s)

3

May 21, 2026

VBench Long (test)

63.82Semantic Fidelity

3

May 15, 2026

0.763Dynamic Degree

3

May 12, 2026

83.69Total Score

3

Apr 24, 2026

64.37Aesthetic Quality

3

Feb 26, 2026

CineLOG 1.0 (test)

93.5Genre Correctness

3

Feb 26, 2026

T2V-CompBench 640 × 480p fixed at ξ = 0.85 Wan2.2

0.843Consistency

3

Feb 26, 2026

VBench and Movie Gen Bench (user study)

44.4Prompt Alignment

3

Feb 26, 2026

User Study 90 videos

66.5Win Rate

3

Feb 26, 2026

65 dynamic scene prompts

76.3Photorealism Score

3

Feb 26, 2026

7.64FID

3

Feb 26, 2026

73.61Rank

3

Feb 26, 2026

Dynamic-Bench (test)

HunyuanVideo-1.5 + SpecLoR-Zero

3.283Text Alignment Score

2

Jun 11, 2026

Page 8 of 9

Popular datasets

VBench Simulated Videos Wan HunyuanVideo User Study

Follow for updates

Join SOTA2 Contact