MiniMax H3 shows near-Seedance-2.0 music generation with Spanish lyrics
kornerson · reddit · 2026-08-04
MiniMax H3 is being used to generate music and songs, and the poster says it feels close to Seedance 2.0 quality.
- The example uses tropical music and Spanish lyrics.
- The main claim is that the model has a strong understanding of prompts.
- This is presented as a fun, creative demo rather than a technical benchmark.
More from Multimodal
- StemFX frames music mixing as token prediction on 105K songs — affige_yang · 2026-08-04
- A Reddit user turned Minimax H3 into a 2D Pokémon battle animation — chille9 · 2026-08-04
- Cover-song generation gets more reliable when phoneme timing is extracted — affige_yang · 2026-08-04
- A repost teases a 30-second video from an upcoming model, not Seedance 2.5 — iamaliveix · 2026-08-04
- ISMIR 2026 paper MPEcho adds phoneme conditioning to cut lyric errors in cover song generation — affige_yang · 2026-08-04
- Opus 5 prompt turns an android-body romance into a Suno song — almostsweet · 2026-08-04