Lyria 3.5 music model reaches the Gemini API and the Gemini app

Archive entry: this covers an event from September 3, 2026, and we published it on November 8, 2026. The details come from sources dated at the time and may have changed since.

On September 3, 2026, Google made lyria-3.5 generally available in the Gemini API, according to the Gemini API changelog. The music model generates full-length songs with improved coherence, natural vocals and control over duration and structure.

What changed

• Gemini API: lyria-3.5 became generally available on September 3, 2026.
• Inputs and output: the model accepted text and image input and produced 44.1 kHz stereo audio.
• Gemini app: a Google blog post dated September 4, 2026 announced Lyria 3.5 in the Gemini app for all users globally, on web and mobile.
• App controls: Gemini app users could choose a genre, vocals or instrumental, templates, and short or long tracks.
• Other places: Lyria 3.5 was also available in Flow Music, Google AI Studio and Google Vids.

What it means for people who build with AI

Lyria 3.5 was open to developers through the API and to every Gemini app user. For people building products, the API combined text and image input with 44.1 kHz stereo output and control over the duration and structure of songs.

An earlier launch

According to the Google DeepMind blog listing, Lyria 3.5 had already launched in Flow Music in July 2026, before the API and app release.

Sources

• Gemini API changelog: September 3, 2026 (Google AI Studio (Gemini API changelog), September 3, 2026, official)
• Create your best tracks yet with Lyria 3.5 in Gemini (Google (The Keyword), September 4, 2026, official)
• Google DeepMind blog listing, page 2 (Google DeepMind, official)

Written with AI assistance, strictly from the sources listed above. Before it went live, automated checks verified its dates, sources and every language edition.