# Lyria 3 Pro Preview

> Lyria 3 Pro Preview is a proprietary language model from Google DeepMind, released on 25 Mar 2026. It accepts text and images and generates text and audio, with a 1M-token context window and up to 66K output tokens per response.

Source page: https://www.aimodel.directory/models/lyria-3-pro-preview
Last verified: 4 Oct 2026

## Specifications

- **Developer:** Google DeepMind
- **Release date:** 25 Mar 2026 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Weights:** Proprietary (API only) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Context window:** 1M (1,048,576 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Max output:** 66K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Input:** Text and images (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Output:** Text and audio (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Reasoning mode:** No (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Tool calling:** No (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)
- **Structured output:** No (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/google/models/lyria-3-pro-preview.toml)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| Gemini API | Free | Free | – | 1M |
| OpenRouter | Free | Free | – | 1M |

## FAQ

### What is the context window of Lyria 3 Pro Preview?

Lyria 3 Pro Preview has a context window of 1,048,576 tokens (1M) and can generate up to 65,536 tokens in a single response, according to models.dev.

### Is Lyria 3 Pro Preview open source?

No. Lyria 3 Pro Preview is a proprietary model; Google DeepMind has not released its weights. It is available through APIs including Gemini API, OpenRouter.

### When was Lyria 3 Pro Preview released?

Google DeepMind released Lyria 3 Pro Preview on 25 Mar 2026, according to models.dev.

### Which providers offer Lyria 3 Pro Preview?

We track Lyria 3 Pro Preview on 2 providers: Gemini API, OpenRouter.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "Lyria 3 Pro Preview", https://www.aimodel.directory/models/lyria-3-pro-preview