model release

Google releases Lyria 3 Clip Preview for music generation via API

TL;DR

Google has released Lyria 3 Clip Preview, a music generation model available through the Gemini API as of March 30, 2026. The model generates 30-second audio clips from text prompts or images at $0.04 per clip, with a 1,048,576 token context window.

2 min read
0

Google Releases Lyria 3 Clip Preview Music Generation Model

Google has launched Lyria 3 Clip Preview, a music generation model available through the Gemini API starting March 30, 2026. The model generates short audio clips, loops, and previews from text prompts or images.

Key Specifications

Context Window: 1,048,576 tokens

Pricing: $0.04 per 30-second audio clip. Input and output token pricing is listed as $0/M, suggesting the per-clip pricing model supersedes traditional token-based billing.

Audio Quality: The model generates high-quality, 48kHz stereo audio with structural coherence, including vocals, timed lyrics, and full instrumental arrangements.

Capabilities

Lyria 3 Clip is part of Google's broader Lyria 3 family of music generation models. It can:

  • Generate audio from text prompts
  • Generate audio from images
  • Produce clips up to 30 seconds in duration
  • Create loops and preview content
  • Generate vocals with synchronized lyrics
  • Arrange full instrumental compositions

Availability

The model is accessible through the Gemini API and is routed through OpenRouter, which handles provider selection and fallback management for uptime optimization. Developers can access the model using OpenAI-compatible APIs or through the OpenRouter SDK.

Usage data shows current demand, with prompt activity at 410K and completion activity at 280K tokens tracked across the platform in recent monitoring periods.

What This Means

Google enters the consumer music generation market with a pricing model that simplifies billing compared to token-based systems—developers pay per 30-second clip rather than tracking input/output token consumption. The 1M+ token context window is a technical feature that likely supports longer creative instructions or batch processing capabilities, though typical use cases focus on discrete clip generation. This positions Lyria 3 Clip as a tool for music preview generation, loop creation, and short-form content production, competing with existing music AI tools at a transparent, clip-based price point.

Related Articles

model release

NVIDIA Releases Nemotron 3.5 Lightning: 30B MoE Model with 1M Token Context and 3B Active Parameters

NVIDIA released the full-precision BF16 reference weights for Nemotron 3.5 Lightning, a 30B-parameter Mixture-of-Experts model with only 3B active parameters and support for up to 1 million tokens of context. The model uses a hybrid Mamba-2, MoE, and Attention architecture and is licensed under OpenMDW-1.1 for commercial use.

model release

Qwen Releases Qwen3.8 2.4T A95B, a 2.4-Trillion-Parameter Open-Weight MoE Model

Qwen has released Qwen3.8 2.4T A95B, an open-weight sparse mixture-of-experts model with 2.4 trillion total parameters and 95 billion active parameters per forward pass. The model is the open-weight variant of Qwen3.8 Max, targeting coding, research, complex reasoning, and agentic workflows with a 262K token context window.

model release

xAI Releases Grok 4.6, a 1.5T-Parameter Model Powering New 'Grok Bot' AI Teammate Product

xAI released Grok 4.6, a confirmed 1.5T-parameter model built on Grok 4.5 with heavier training on long-horizon agentic tasks. It powers the newly launched Grok Bot product and scores 61 on Artificial Analysis's Intelligence Index at $2/$6 per 1M input/output tokens — well below frontier competitors.

model release

Alibaba Releases Qwen3.8-2.4T-A95B-FP8: 2.4T-Parameter Open Model with 1M-Token Context

Alibaba's Qwen team has released Qwen3.8-2.4T-A95B-FP8, an open-weight, FP8-quantized MoE model with 2.4 trillion total parameters and 95 billion activated per token. It natively supports 262,144 tokens of context, extensible to 1,010,000, and forms the base for the hosted Qwen3.8-Max API.

Comments

Loading...