Text-to-Video Generation on VBench 100 sampled prompts benchmark leaderboard · SOTA2 Research

Text-to-Video Generation on VBench 100 sampled prompts

284.71HD-FVDAtlasVid

Evaluation Results

Method	Links
AtlasVid2026.05		284.71	0.6695	0.3019	97.41	96.54	99.57	99.31	82.21	63.31	68.52	98.31
T3-Video2026.05		311.29	0.6471	0.2972	97.02	96.23	95.12	99.17	78.92	62.96	68.17	97.42
UltraWan2026.05		372.39	0.6211	0.2903	95.71	97.94	98.86	99.06	62.22	59.52	67.39	99.1
UltraGen2026.05		432.71	0.6358	0.2848	94.19	96.33	98.9	98.72	71.34	58.13	65.88	97.13
Wan2.12026.05		512.32	0.6424	0.3076	96.34	97.29	99.49	97.44	85.56	62.43	66.51	89.58