Suno retires older models: keeping a brand jingle reproducible
TechCrunch says Suno will retire older models after v6. A brand jingle needs a stored master file and a saved prompt, not a promise to regenerate it.

If a brand jingle was made on an older Suno model, plan on never regenerating it. TechCrunch reports that Suno's new v6 models replace the old ones and that older models are to be retired. Even on a model that stays available, generative music is not repeatable: Google's Lyria guide says outputs vary across identical calls. The only reliable copy of a jingle is the file you saved, together with the record of how it was made.
What is known about model lifetimes
Two facts, from two sources, point the same way.
| Source | Statement | Implication |
|---|---|---|
| TechCrunch on Suno v6 | Older models are to be retired; v6 models were not trained on earlier data | The model that made your jingle may stop being available |
| Google Gemini API guide | Outputs vary across identical calls; single-turn only | Even a live model will not give you the same track again |
| Sume Music docs | No seed, temperature, guidance or duration parameter | There is no setting that makes generation repeatable |
What to store with a jingle
Treat the generated file as the master and keep these next to it.
- The delivered audio file itself, in lossless form where the tool offers it. Google's guide lists WAV alongside MP3 for Lyria.
- The exact prompt, including any section tags or timestamps.
- The model id that ran, and the date.
- The job or generation id, and the plan or terms in force the day you generated it, which matters if rights depend on a paid tier.
- Any stems or edits made later, labelled as derived from the master.
How this looks in Sume
Sume's Music Router returns completed audio as a hosted artifact on Sume media URLs, and the job records request.routed_model, the engine that ran, such as lyria-3.5. You can pin that id instead of using sume/music-auto when you want the engine to stay fixed across calls. Pinning chooses the engine; it does not make the output repeatable, so the stored artifact is still the master.
Download the artifact into your own asset store soon after the job completes. The docs describe hosted artifacts but do not set a retention promise, so I would not rely on a link as an archive. Sume does not ship Suno; this checklist applies to any music tool.
If you need a jingle to evolve
When the brief changes, such as a new length or a different mood, write a new prompt from the saved one and generate again, accepting that it is a new piece. For continuity, keep the prompt's tempo, key and instruments the same and name the same arc. Sume's music docs advise keeping continuity when you want one consistent score, and an optional still can go in as image_url. Listen to every result; a similar brief gives a related track, not the same one.
Sources
Related posts
More in Use cases
- Swap the person in a clip: reference video or H3 Max Recast on Sume
Seedance 2.5 offers reference-based editing. For a straight person swap, Sume lists H3 Max Recast. When to use which, and the inputs each needs.
- Replace audio from 12.4 s to 14 s of a clip with Timeline parts
Seedance 2.5 lists timestamp-level audio editing. On Sume you can detach a clip's audio and rebuild it from parts with a replaced line in the middle.
- Swap a product's colorway in a finished clip: Gemini Omni edit
One approved ad, five colorways: gemini-omni-flash-1.1 on Sume edits a source video from a prompt. Input rules, what it cannot carry, and a batch plan.
- SynthID not detected in Gemini: that does not mean the video is real
Gemini's checker only recognizes Google AI content, so no watermark means no Google mark was found, not that a person filmed it. Know its limits.
Written by Sume