MiniMax Music 3.0
MiniMax Music 3.0 is a music-generation API for creating songs from a description, lyrics, or reference audio. Developers use the prompt parameter to define style, mood, instrumentation, vocal character, and production direction, while the lyrics parameter supplies vocal content. Its upgraded semantic model improves creative-intent understanding and reduces drift in AI-generated music. Higher sound quality produces clearer mixes and supports specific instruments and playing techniques such as slides and legato. A new vocal engine delivers more natural synthesis with control over melody, pronunciation, breathing, and layered harmonies. Teams can first call the Lyrics Generation API to write full lyrics with sections such as Verse, Chorus, and Bridge, then send them to the Music Generation API, or skip that step and generate a song directly with lyrics optimization. Music 3.0 also supports instrumental-only creation.
Learn more
MiniMax H3
MiniMax H3 is a general-purpose omni-modal generation model that jointly understands multimodal contexts spanning text, images, video, and audio. It generates videos with native stereo sound at up to 2K resolution and 15 seconds in length, delivering content for advertising, branding, ecommerce, product design, UI/UX, gaming, and creative workflows. Users can combine reference types in one instruction, for example, transferring camera movement from a video, placing a character from an image into the scene, and matching vocals from an audio clip, while describing the relationships in natural language. H3 supports text-to-image, text-to-video with jointly generated audio, multi-shot modeling, text-to-audio, and generalized reference and editing across images, videos, and audio. Voice, sound effects, and music are modeled together. The model excels at instruction following, accurate text and brand presentation, and video-to-video motion transfer.
Learn more
Dessa
Dessa is Block's speech-to-speech artificial intelligence model for conversational voice research and applications. It takes audio input and produces audio output directly, without converting speech into text as an intermediate step. Trained from scratch on human conversation rather than text, the model supports full-duplex interaction, listening and speaking at the same time, and runs locally on a MacBook Pro with an M4 Max chip or better. It captures conversational cues such as emotion, sarcasm, laughter, pauses, and turn-taking that can be lost in text-based pipelines. Dessa is intended for AI safety researchers and developers studying voice-model behavior, privacy, and real-time interaction. It is a speech model for research and development, not a general-purpose business application or hosted voice-agent platform.
Learn more
MusicGPT
MusicGPT is an AI-powered music creation platform that lets you generate full original music, beats, instrumentals, lyrics, vocals, sound effects and soundscapes simply by typing a description of what you want, letting the AI produce professional quality tracks across genres in seconds. It provides tools to edit audio, upload and transform existing files, extract stems, remix tracks or create sound effects and samples with hyper-realistic quality, and explore a royalty-free music library for discovery and inspiration. It includes a simple prompt box for song creation, support for text-to-speech with thousands of realistic voices, an AI voice changer, AI stem splitter, audio enhancements and the ability to isolate vocals or instruments. MusicGPT runs on proprietary AI audio technology and integrates via a flexible API for developers to power apps or projects, while users can stream and download unlimited music they create.
Learn more
