Google DeepMind

Lyria 3.5

Generally availableProprietary131K context

Summary

Lyria 3.5 is a proprietary language model from Google DeepMind, released on 3 Sept 2026. It accepts text and images and generates text and audio, with a 131K-token context window and up to 66K output tokens per response.

Music generation model for full-length songs with vocals and structure from text or image prompts

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperGoogle DeepMind
Release date3 Sept 2026 [models.dev]
StatusGenerally available [models.dev]
WeightsProprietary (API only) [models.dev]
Context window131K (131,072 tokens) [models.dev]
Max output66K tokens [models.dev]
InputText and images [models.dev]
OutputText and audio [models.dev]
Reasoning modeNo [models.dev]
Tool callingNo [models.dev]

Lyria 3.5 API pricing by provider

USD per million tokens. Sorted by input price.

We do not currently track a paid API listing for this model.

Frequently asked questions

What is the context window of Lyria 3.5?

Lyria 3.5 has a context window of 131,072 tokens (131K) and can generate up to 65,536 tokens in a single response, according to models.dev.

Is Lyria 3.5 open source?

No. Lyria 3.5 is a proprietary model; Google DeepMind has not released its weights.

When was Lyria 3.5 released?

Google DeepMind released Lyria 3.5 on 3 Sept 2026, according to models.dev.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
GPT-Live 1OpenAI10 Sept 2026128K––Closed
Arrow 2Quiverai16 Sept 2026131K$4$20Closed
Arrow 2 TelosQuiverai16 Sept 2026131K$6$30Closed
Toast 1Mixedbread13 Aug 2026131K$0.30$0.72Closed
Embed 5 FastCohere30 Sept 2026128K––Closed
Embed 5 ProCohere30 Sept 2026128K––Closed

Change history

  • Added to the directory

Sources