List your company

Text-to-Music Generation on MusicCaps

1.01KLDMusicLM
0.98681.14341.31.4566Jan 26, 2023Aug 19, 2023Mar 11, 2024Oct 2, 2024Apr 25, 2025Nov 16, 2025Jun 10, 2026

Evaluation results

MethodLinks
MusicLM2023.011.010.4440.51312———————
TangoMusicTask=T2M2026.061.13————1.882.86—155.577.060.23
Riffusion2023.011.190.7613.40.34158———————
AudioLDM 2 (Full)Params=712M, Hours=17.9k2024.091.2————3.13——————
MusicGenParams=1.5B, Hours=20k2024.091.22————3.8—0.31————
FluxMusicParams=2.1B, Hours=22K2024.091.25————1.432.980.36————
AudioLDM-2-LargeTask=T2M2026.061.26————2.82.84—15.615.226.70.23
Jen-1Params=746M, Hours=5.0k2024.091.29————2.11—0.33————
QA-MDT (DiT)Params=675M, Hours=12.5k2024.091.31————1.652.80.35————
AudioX-TurboTask=T2M2026.061.31————1.543.61—9.54.896.550.22
MAGNET-largeTask=T2M2026.061.32————4.241.98—23.885.846.710.19
AudioX-BaseTask=T2M2026.061.38————1.673.47—12.894.576.450.2
MusicGenTask=T2M2026.061.43————4.552.24—25.45.197.160.18
AudioLDM-L-FullTask=T2M2026.061.45————6.342.49—34.444.726.10.22
AudioLDM 2 (Music)Params=712M, Hours=10.8k2024.091.46————4.042.670.34————
QA-MDT (U-Net)Params=1.0B, Hours=12.5k2024.091.51————2.032.410.33————
Stable Audio OpenTask=T2M2026.061.51————3.232.94—36.333.917.180.23
Mubert2023.011.580.459.60.3297———————
MousaiParams=1042M, Hours=2.5k2024.091.59————7.5—0.23————
Mubert2023.02————254———————
MusicCapsType=Ground Truth Reference2023.01————472———————
MusicCaps (Ground Truth)2023.02————959———————
MusicLM2023.02————692———————
MusicLMParams=1290M, Hours=280k2024.09—————4——————
Noise2MusicArchitecture=Waveform2023.02————718———————
Riffusion2023.02————308———————