Lyria Models by Google DeepMind
Google DeepMindProprietary
Google DeepMind releases · 50 in the last 12 months · this family litChangelog →
6 models2024–2026Up to 131k ctx
Last refreshed 2026-09-03. Next refresh: weekly.
Details
ResearcherGoogle DeepMind
LicenseProprietary
Commercial useCommercial use: conditional
Models6
Released2024–2026
Max context131k
Capabilities
Vision4 of 6 models
Multimodal4 of 6 models
About
Google DeepMind's Lyria family of music and audio generation models. Creates high-quality music tracks, audio clips, and real-time audio from text prompts.
Decision facts
- Best fit
- audiovision and multimodal workstructured outputs
- Capability starting point
- Lyria 3.5 Clip Preview with 131k context and multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- DiffusionGemma
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
6 in view
Lyria 3.5 Clip PreviewCurrent
Use when the workload needs audio, 131k context, and multimodal inputs.
2026-09audio131k contextmultimodal inputs
Lyria 3.5 Pro PreviewCurrent
Use when the workload needs audio, 131k context, and multimodal inputs.
2026-09audio131k contextmultimodal inputs
Lyria 3 ClipCurrent
Use when the workload needs audio and multimodal inputs.
2025-01audiomultimodal inputs
Lyria 3 ProCurrent
Use when the workload needs audio and multimodal inputs.
2025-01audiomultimodal inputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Lyria 3.5 Clip Preview | Use when the workload needs audio, 131k context, and multimodal inputs. | 2026-09 | audio131k contextmultimodal inputs | Current |
| Lyria 3.5 Pro Preview | Use when the workload needs audio, 131k context, and multimodal inputs. | 2026-09 | audio131k contextmultimodal inputs | Current |
| Lyria 3 Clip | Use when the workload needs audio and multimodal inputs. | 2025-01 | audiomultimodal inputs | Current |
| Lyria 3 Pro | Use when the workload needs audio and multimodal inputs. | 2025-01 | audiomultimodal inputs | Current |
| Lyria RealTime | Use when the workload needs audio. | 2024-11 | audio | Current |
| Lyria 2 | Use when the workload needs audio. | 2024-06 | audio | Current |
Release Timeline
4 release groups2026-09
2 current
Lyria 3.5 Clip Preview
Currentaudio131k contextmultimodal inputs
Lyria 3.5 Pro Preview
Currentaudio131k contextmultimodal inputs
2025-01
2 current
Lyria 3 Clip
Currentaudiomultimodal inputs
Lyria 3 Pro
Currentaudiomultimodal inputs
2024-11
1 current
Lyria RealTime
Currentaudio
2024-06
1 current
Lyria 2
Currentaudio
Specifications(6 models)
| Model | Released | Context | Vision | Multimodal |
|---|---|---|---|---|
| Lyria 3.5 Clip Preview | 2026-09 | 131k | Yes | Yes |
| Lyria 3.5 Pro Preview | 2026-09 | 131k | Yes | Yes |
| Lyria 3 Clip | 2025-01 | — | Yes | Yes |
| Lyria 3 Pro | 2025-01 | — | Yes | Yes |
| Lyria RealTime | 2024-11 | — | No | No |
| Lyria 2 | 2024-06 | — | No | No |
