MiniMax Releases Free Music Generation AI MiniMax-Music3 With Japanese Vocals
MiniMax-Music3 extends the company's open-model push into music generation, offering free local generation of Japanese vocal songs up to five minutes long.
Reporting from 1 source: GIGAZINE.
Chinese AI developer MiniMax released the music generation AI MiniMax-Music3 on August 14, 2026. The model generates songs with vocals from text prompts, supports Japanese vocals, and can produce tracks up to five minutes long. It is free to use and can run locally, with a demo app available on Hugging Face.
MiniMax released MiniMax-Music3 on August 14, 2026, Japan time. The model generates songs with vocals from text instructions and supports Japanese vocal generation. It can produce tracks up to five minutes long and outputs 32kHz 16-bit stereo music files.
The model is free and can run locally. A demo app on Hugging Face lets users generate songs in a browser, though the free tier limit was reached after one one-minute song during testing. Local generation through ComfyUI is unlimited. MiniMax also released an AI skill that writes long lyrics and structured prompts from short user input.
MiniMax has announced plans to release image generation and editing models based on MiniMax H3, plus video upscaling models, as open models in the future.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.