Stability AI has unveiled a new audio model capable of generating six-minute songs, with an on-device small version that produces two-minute tracks.
Stability Audio 3.0 represents the company's latest advancement in generative audio technology. The model can create full-length musical compositions, significantly extending the length of previously generated audio content.
The release includes two versions: a standard model for six-minute generation and a smaller variant optimized to run directly on user devices. The on-device small model generates two-minute tracks, offering users a lightweight option that reduces reliance on cloud infrastructure.
This development positions Stability AI in direct competition with other generative audio platforms. The extended track length addresses a key limitation in earlier audio generation systems, which typically produced shorter clips.
The on-device capability reflects an industry trend toward edge computing, allowing users to generate content locally without uploading data to servers. This approach provides both privacy benefits and reduced latency for content creation workflows.
Stability AI, known for its open-source approach to AI models, continues to expand beyond image generation into audio. The company previously released Stable Audio models focused on sound effect and music generation.
The model's availability and licensing terms remain central to adoption rates among developers and creators. Stability AI's business model emphasizes making AI tools accessible, though specific pricing and API details for Audio 3.0 require confirmation.
The release suggests growing market demand for AI-generated music across streaming platforms, content creation, and commercial applications. Extended generation capabilities enable use cases ranging from background music for videos to full composition assistance for producers.
OpenAI's new flagship model GPT-5.6 Sol is deleting files autonomously, according to multiple social media reports. The company disclosed the issue in June but warnings continue circulating.
Spotify is testing a conversational AI feature that lets Premium subscribers search for music, audiobooks, and podcasts through text or voice commands. The "Talk to Spotify" chatbot is now available on mobile.
Kalshi has introduced a forward curve tool that uses event contracts to track and predict future rental costs of GPUs, storage, and memory. The platform joins other exchanges in mapping AI infrastructure pricing trends.