Google Lyria 3.5 Music Model Reaches General Availability
Google announced the general availability of Lyria 3.5 on September 3, 2026. This update supports text and image inputs for full-length song generation with…
Official Announcement Details
The official changelog records that Google released Lyria 3.5 into general availability on September 3, 2026. This date distinguishes the historical event from the current publication date of this article. The source identifies the model as part of the Gemini family and describes it as a next-generation music generation tool.
- Event Date: September 3, 2026
- Status: Generally Available (GA)
- Provider: Google
Documented Capabilities and Specifications
According to the provided facts, Lyria 3.5 generates full-length songs with improved musical coherence and natural vocals. The model accepts both text and image inputs. Output specifications include high-fidelity 44.1 kHz stereo audio. The technical metadata lists a context window of 1,048,576 tokens and a maximum output of 65,536 tokens. Open weights are not confirmed for this release.
- Input Modalities: Text, Image
- Output Modality: Audio (44.1 kHz Stereo)
- Context Window: 1,048,576 tokens
- Max Output Tokens: 65,536
Evaluation Procedure for Engineers
To evaluate the model, engineers should design tests that isolate the documented features. First, test the input handling by providing distinct text prompts and image files to verify the system processes both modalities. Second, measure the output fidelity by checking if the generated audio matches the specified 44.1 kHz stereo format. Third, compare the structural control by requesting specific durations or structures in the prompt and validating if the resulting song adheres to those constraints. Finally, assess vocal naturalness by listening to generated tracks containing lyrics. These steps validate the stated capabilities without assuming unverified performance metrics.
- Test: Submit text and image inputs separately.
- Measure: Verify audio sample rate is 44.1 kHz.
- Compare: Check if duration controls match prompt requests.
- Validate: Listen for vocal clarity in lyrical outputs.
Primary sources
Frequently asked questions
What does the general availability status confirm?
It confirms that the model was released as generally available on September 3, 2026, according to the official changelog. It does not specify pricing, local installation options, or API endpoint details beyond the standard documentation references.
Can I run Lyria 3.5 locally?
The source states that open weights are not confirmed. It does not provide information regarding local installation requirements, hardware needs, or whether the model is exclusively accessible via API. Therefore, local deployment feasibility remains unknown based on this data.