ResearchBenchmarksAudio-Visual Generation on Verse-Bench multi-speakerAudio-Visual Generation on Verse-Bench multi-speakerFollow14.9cpCERMOVA13.75221.50129.2536.999Feb 9, 2026Evaluation resultsMethodMethodLinkscpCERMOVAResolution=720p, s_B=1...Resolution=720p, s_B=1.0, dual CFG=false2026.0214.9MOVAResolution=360p, s_B=1...Resolution=360p, s_B=1.0, dual CFG=false2026.0217.7MOVAResolution=720p, s_B=3...Resolution=720p, s_B=3.5, dual CFG=true2026.0221.8LTX-2Resolution=720pResolution=720p2026.0222MOVAResolution=360p, s_B=3...Resolution=360p, s_B=3.5, dual CFG=true2026.0224.7OviResolution=720pResolution=720p2026.0243.6