Google: Lyria 3 Pro Preview

Google's Lyria 3 Pro Preview is listed as accepting text and image input and producing text and audio output, with a one-million-token context window. No benchmark results, pricing, or tool-use information are included. As a result, the supplied data does not establish its reasoning performance, audio quality, or readiness for reliable production use.

Its audio-output modality makes it a possible candidate for creative workflows that combine prompts or images with audio generation. But the data does not specify controllability, latency, or whether outputs meet client delivery standards. The preview designation and lack of measured results make this an exploration option rather than a dependable general-purpose model. The context window is a technical feature, not evidence of quality.

Evaluate it only for audio-centric experiments, and compare actual outputs against creative briefs before adopting it. Production use should wait on demonstrated quality and clear cost information.

Assessed September 23, 2026

The DFO score ranks models on the Artificial Analysis Intelligence Index and is refreshed as new models launch. The write-up and verdict are our own. Feedback?