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

Video Generation research benchmarks – Page 22 · SOTA2 Research

Video Generation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

HECTOR Multi-Object (Evaluation Set)

0.695R-CLIP Score

3

Mar 11, 2026

55.1Overall Preference Score

3

Mar 10, 2026

Human Evaluation win rate

50.31Quality Win Rate

3

Mar 6, 2026

VBench Out-of-domain

83.98Semantic Fidelity

3

Feb 26, 2026

Wan In-domain 2.1

3.55VQ Score

3

Feb 26, 2026

91.64Anatomy

3

Feb 26, 2026

Adobe Stock-220 20s clips (test)

80.15Average Score

3

Feb 26, 2026

VBench long-video (60s)

97.86Temporal Flickering

3

Feb 26, 2026

Animate3D (test)

0.945I2V Score

3

Feb 26, 2026

Diffusion4D (test)

0.933I2V (Fidelity)

3

Feb 26, 2026

EPIC-Kitchens (Unseen)

Mitty (T2V 14B)

9.35FVD

3

Feb 26, 2026

EPIC-Kitchens (Seen)

Mitty (T2V 14B)

6.9FVD

3

Feb 26, 2026

Human2Robot (test)

Mitty (T2V 14B)

6.48FVD

3

Feb 26, 2026

I2VBench 24 frames

97.3Subject Consistency Score

3

Feb 26, 2026

DL3DV 5 fps (test)

61.18FID

3

Feb 26, 2026

5Video Length (s)

3

Feb 26, 2026

GSCollision Novel-View (NV)

0.49VLM PhysR

3

Feb 26, 2026

GSCollision Novel-Background (NB)

56VLM PhysR Score

3

Feb 26, 2026

GSCollision Compositional

0.47VLM PhysR Score

3

Feb 26, 2026

VBench Wan2.1 1.3B extrapolation, 141x448x832 1.0 (test)

64.27AQ

3

Feb 26, 2026

VBench Wan2.1 1.3B extrapolation, 61x720x1280 1.0 (test)

65.49Aesthetic Quality

3

Feb 26, 2026

0.051MEt3R

3

Feb 26, 2026

61.4DINO Score

3

Feb 26, 2026

Human evaluation set 15 videos (test)

4.4Image Prompt Alignment

3

Feb 26, 2026

YouTube Driving

194.8FVD (16 frames)

3

Feb 26, 2026

Page 22 of 24

Popular datasets

VBench Wan User Study VideoPhy RealEstate10K

Follow for updates

Join SOTA2 Contact