Text-driven Motion Generation on Motion-X (test) benchmark leaderboard · SOTA2 Research

Text-driven Motion Generation on Motion-X (test)

0.595R Precision Top 1Real motions

Evaluation Results

Method	Links
Real motions2026.02		0.595	0.787	0.868	0	3.717	—	0.672
LG-Tok-mid2026.02		0.591	0.781	0.864	0.076	3.758	2.245	0.682
LG-Tok-mini2026.02		0.588	0.776	0.857	0.071	3.835	2.235	0.681
LG-Tok2026.02		0.582	0.775	0.858	0.088	3.844	2.294	0.682
MotionDiffuse2026.02		0.559	0.752	0.839	0.954	4.167	2.476	0.66
MARDM2026.02		0.528	0.727	0.82	0.147	4.433	3.077	0.643
MoSa2026.02		0.513	0.703	0.796	0.21	4.783	3.167	0.654
MoMask2026.02		0.502	0.694	0.79	0.247	4.832	2.715	0.644
StableMoFusion2026.02		0.474	0.682	0.787	0.213	4.888	2.985	0.607
T2M-GPT2026.02		0.47	0.644	0.735	1.085	5.488	12.807	0.622
MMM2026.02		0.424	0.6	0.695	2.918	6.098	2.342	0.607