ResearchBenchmarksText to Audio-Visual Generation on Audio-Visual Generation Benchmark (test)Text to Audio-Visual Generation on Audio-Visual Generation Benchmark (test)Follow4.63VALTX-23.23643.59823.964.3218May 9, 2026Evaluation resultsMethodMethodLinksVAIDPQCUWERTATVAVLSE-CLSE-DDSLTX-2Input Modality=T2AVInput Modality=T2AV2026.054.634.686.125.720.113.7426.35813.627.757UnisonInput Modality=T2AVInput Modality=T2AV2026.054.514.596.175.780.093.6226.21863.557.956OviInput Modality=T2AVInput Modality=T2AV2026.054.224.516.085.650.183.5525.99833.478.058JavisDiTInput Modality=T2AVInput Modality=T2AV2026.053.294.524.833.731.813.5324.31491.81—53