Lyria 3 Pro Preview

Google· released 30 Mar 2026

google/lyria-3-pro-preview
Input / 1M tokens
free
Output / 1M tokens
free
Context window1M tokens

Longest answer: 66K tokens

reads textreads imagevision

About

Lyria 3 Pro Preview is a Google model in the Lyria 3 family of music generation models, available through the Gemini API. Full-length songs are priced at $0.08 per song. The listed model accepts text and images, and its outputs are text and audio. Its model ID is `google/lyria-3-pro-preview`; it was released on 2026-03-30. A 1,048,576-token context window and maximum output of 65,536 tokens are listed. The entry also lists free access and vision support. The modality field is text, while the outputs include audio. For builders evaluating a music-generation option in an existing stack, the Gemini API availability, listed song price and text-and-image inputs are relevant details. The preview designation is part of the model name; no further preview behavior is specified here.

Who it is for

It may suit teams exploring music generation through the Gemini API, particularly where text or image inputs and audio output fit the workflow. The listed price is $0.08 per full-length song.

What is good

  • Full-length songs cost $0.08 per song
  • Available through the Gemini API
  • Accepts text and image inputs
  • Outputs text and audio

What to know first

  • Preview model designation
  • Modality is listed as text

Verdict

Lyria 3 Pro Preview offers audio output as part of Google's Lyria 3 music model family, with Gemini API availability and a per-song price. Its preview designation and listed modality details are worth noting when evaluating fit.

Details

Lab
Googleopenrouter.ai · 3 Oct 2026
Context
1,049Kopenrouter.ai · 3 Oct 2026
Input price
Freeopenrouter.ai · 3 Oct 2026
Output price
Freeopenrouter.ai · 3 Oct 2026
Max output
65,536 tokensopenrouter.ai · 3 Oct 2026
Inputs
text, imageopenrouter.ai · 3 Oct 2026
Open weights
Noopenrouter.ai · 3 Oct 2026

More longest-context models

See the list