Google: Lyria 3 Clip Preview

google/lyria-3-clip-preview · GoogleNot yet established

Declared family: Unknown
Entered catalogue: 30 March 2026
Last tested: 3 September 2026
Evidence confidence: Unknown

Identity statement

Google: Lyria 3 Clip Preview has a clean measured fingerprint that currently sits alone: no other measured model shares its signature.

What we checked

Four different kinds of evidence, and what each one showed

We checked	What we found	Where it came from
How it counts tokens	Counted, using the 49 strings that came back reliably	we measured it
How its API is set up	Set up identically to 1 other listing, field for field	api_4e7bfe54
How much it can read and write	Reads up to 1,048,576 tokens at once, writes up to 65,536	its own listing
Who served our requests	Google AI Studio	we measured it, 3 September 2026

Closest measured models

Agreement across the mutually clean test strings

Poolside: Laguna S 2.1poolside/laguna-s-2.15/49 · 10%Meta: Llama Guard 4 12Bmeta-llama/llama-guard-4-12b5/49 · 10%OpenAI: GPT-3.5 Turbo Instructopenai/gpt-3.5-turbo-instruct4/40 · 10%Reka Flash 3rekaai/reka-flash-34/45 · 9%Anthropic: Claude Sonnet 4.6anthropic/claude-sonnet-4.64/48 · 8%Anthropic: Claude Opus 4.6anthropic/claude-opus-4.64/48 · 8%

Click a row to open the full comparison.

The measured fingerprint

What each test string cost this model, in tokens

We sent Google: Lyria 3 Clip Preview fifty short pieces of text and recorded what each one cost it in tokens. The bars below are those costs. Two models built on the same tokenizer produce the same bars; a model built on a different one produces a different set, which is what makes this a fingerprint.

All 50 rows, 1 excluded as corrupt

English and whitespace

en-prose17

en-long24

spaces-207

spaces-6017

tabs-2012

newlines-2012

mixed-ws5

Digits

digits-93

digits-124

digits-309

digits-sep4

float-longx

CJK

zh-common19

zh-long25

zh-rare10

ja-kana17

ja-kanji15

ko18

Other scripts

ru18

ar33

he25

hi27

th27

el17

Emoji

emoji-basic11

emoji-skin21

emoji-zwj-family12

emoji-zwj-x334

emoji-flags17

emoji-prof20

Rare Unicode

math17

boxdraw18

combining7

cjk-ext-b9

surrogates28

zalgo18

rtl-mix13

Code

py-code17

py-indent21

json16

html12

regex17

camel9

snake11

Repetition and encodings

rare-word-x537

repeat-tok31

base6413

hex9

url15

uuid10

An x marks a row excluded as corrupt (caching or provider interference detected during measurement). Overhead subtracted: 5 prompt tokens.

Declared record

What the catalogue claims about this model

30 second duration clips are priced at $0.04 per clip. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate...

Modalities: text+image->text+audio
Declared tokenizer: Other
Prompt price: $0 / M tokens
Completion price: $0 / M tokens

Supported parameters

max_tokensresponse_formatseedtemperaturetop_p

defaults: {"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null}

Catalogue entry

History

Every observation, kept as taken

30 March 2026Enters the OpenRouter catalogue with no declared family.
3 September 2026Fingerprinted in the YFarmX catalogue sweep · 49 of 50 strings measured clean.