Text-to-Video Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
28.46FID (Image)
5
Feb 26, 2026
98.19Shape Accuracy
5
Feb 26, 2026
96Subject Consistency
4
Jun 18, 2026
0.933Consistency
4
Jun 11, 2026
80Score
4
Jun 4, 2026
1.81Speedup
4
Apr 21, 2026
99.32Temporal Fidelity (T.F.)
4
Apr 15, 2026
98.48T.F.
4
Apr 15, 2026
97.36Temporal Fidelity
4
Apr 15, 2026
867.27FVD
4
Apr 8, 2026
81Human Preference
4
Mar 24, 2026
0.308CLIP Score
4
Mar 19, 2026
3.94Visual Quality
4
Mar 17, 2026
83.75Total Score
4
Mar 9, 2026
29.55PSNR
4
Feb 26, 2026
3.2Generation Speed (x)
4
Feb 26, 2026
3.69User Preference Score
4
Feb 26, 2026
85.5Human Anatomy
4
Feb 26, 2026
83Semantic Objective
4
Feb 26, 2026
76.8Semantic Objective
4
Feb 26, 2026
5.55Physical Plausibility Score
4
Feb 26, 2026
0.082VFR Auto
4
Feb 26, 2026
0.172VFR (Auto)
4
Feb 26, 2026
0.116VFR (Auto)
4
Feb 26, 2026
0.14VFR (Auto)
4
Feb 26, 2026