Flow Music is an AI music creation platform launched by Google Labs. It was formerly ProducerAI (derived from the Riffusion project) acquired at the beginning of the year. It is now integrated into the Google Flow creative ecosystem. Based on Google DeepMind's Lyria 3 Pro music model, it supports the generation of complete songs of up to about 3 minutes, including vocals, lyrics and arrangement, through natural language dialogue.

FlowMusic

Core parameters and statistics

Specific technical parameters (such as model size, context length, supported file formats, input and output restrictions, etc.) are subject to the official product page. It is recommended that users verify the latest technical specifications and system requirements before choosing to ensure that they match their own usage scenarios.

User and market recognition

Gradually build user awareness in the field, and product capabilities are used by content creators and teams to improve work efficiency. Some industry users have incorporated it into their daily workflow. It is recommended to refer to the latest official disclosures for specific user scale and industry adoption rate data.

Cost advantage

Flow Music’s cost logic is to “use AI to significantly reduce the threshold and cost of entire song production.”

• C-side/Individual: Usually a limited free generation quota is provided. Full functions and longer durations require a subscription. Please refer to the official website for details.
• Developer/API: Whether the interface is provided to the outside world is not disclosed.
• Enterprise/Commercial: Commercial authorization and permission must be based on the official page or business communication.

True Cost Structure: Explicit expenses are subscriptions or generation credits, and implicit costs are compliance confirmation for commercial authorization. When evaluating, you should focus on generation quality, controllability, and copyright terms, rather than just "can generate songs."

Main functions

Flow Music’s capabilities are organized around “conversational generation of complete songs”:

• Natural Language Generation: Use dialogue to describe style and theme to generate songs.
• Complete song output: Output the entire song including vocals, lyrics and arrangement, up to about 3 minutes.
• Lyria Model Driver: Based on DeepMind Lyria 3 Pro, with strong sound quality and musicality.
• Flow Ecosystem Integration: Used collaboratively as part of the Google Flow creative ecosystem.

The key to functional value lies in the degree of production: the ability to directly produce usable entire songs is different from tools that only generate fragments.

Model and version evolution

Flow Music's version context is related to its predecessor and underlying model.

Predecessor stage

• ProducerAI (~2025): Derived from the Riffusion project and exists as an independent AI music product.

Integration and upgrade

• Flow Music (~2026, Lyria 3 Pro): After being acquired by Google, it was integrated into Google Flow and upgraded to a music creation platform based on Lyria 3 Pro.

The official unified product version number has not been disclosed, and the version context is expressed in public milestones and model nodes.

Technical advantages

The advantages of Flow Music come from "cutting-edge music model + conversational generation":

Mechanism: Using DeepMind Lyria 3 Pro as the underlying music model, the style, theme, lyrics and other requirements are transformed into a complete arrangement and vocals through dialogue.

Effect: Compared with fragment-level tools, it can directly generate songs with complete structure and vocals, and can be iteratively adjusted through dialogue.

Applicable scenarios: Most suitable for creators and content teams who need to quickly produce complete background music or sample songs.

The price is that the copyright and commercial authorization of AI music are still sensitive issues in the industry, and the terms need to be confirmed before official commercial use.

How to use

Flow Music is available online through the official website:

Usage	Suitable objects	Features
Dialogue generation	Creator	Generate songs using natural language descriptions
Iterative adjustments	Refine requirements	Adjust style and content through dialogue
Flow Ecosystem	Google Flow Users	Used in conjunction with the Creative Ecosystem

The typical process is "describe style and theme → generate complete song → dialogue iteration → export for use". When using it for the first time, it is recommended to test the style and vocal quality with a short description.

Product Pricing

The pricing model is subject to the official real-time page. Usually a freemium or subscription system is used, and basic functions can be used for free. Advanced functions or high-frequency use require paid subscriptions, and users are advised to evaluate the optimal solution based on actual usage.

Application scenarios

• Content Soundtrack: Quickly generate background music for videos and podcasts. The focus of verification is commercial authorization.
• Music Creation Samples: Quickly produce demos or inspired samples, focusing on controllability.
• Personal Creation: Non-musicians generate complete songs, and the focus of verification is the quality of vocals and arrangement.

Applicable people

• Content Creator: Needs original soundtrack but lacks composition ability.
• Music Lovers: Want to quickly turn ideas into complete songs.
• Creative Team: Need a quick demo for proposal or testing.

Unsuitable situations are: professional music production that requires professional-level fine mixing, clear copyright ownership for large-scale commercial use, or that requires complete control over the generated results.

Summary and Outlook

The core value of Flow Music is to use Google DeepMind's Lyria 3 Pro model to directly generate complete songs including vocals, lyrics and arrangements in a conversational manner, which is different from tools that only generate fragments. Google background and model strength are its real advantages.

The current limitation is that the scope of openness and pricing are not yet transparent, and the copyright and commercial licensing of AI music are still common sensitive points in the industry. For users, it is recommended to first use the free credit to verify the style fit and vocal quality before deciding whether to subscribe; before any commercial use, you should focus on confirming the copyright and licensing terms of the generated content.

Related tools: ElevenLabs, udio

Version Info

• Flow Music (Lyria 3 Pro driver) ：The current Lyria 3 Pro music model based on Google DeepMind supports conversational generation of complete songs of up to about 3 minutes. The official version number of the unified product has not been disclosed, and there is no official precise date yet. It is recorded according to the current public form. （~2026-06）
• Formerly ProducerAI (before acquisition) ：Flow Music was formerly known as ProducerAI, which originated from the Riffusion project. After being acquired by Google, it was integrated into the Google Flow ecosystem and upgraded to Flow Music. There is no official precise date yet, it is recorded according to public milestones. （~2025-01）