Snapchat Create Song vs a text-to-song API: Sume Music 1.0 at $0.125

Snapchat's Create Song turns a chat into a song for Lens+ users. Sume Music 1.0 is a text-to-song API: prompt up to 5000 characters, $0.125 per generation.

5 min readSume
All posts

Snapchat's Create Song is a consumer feature inside Lens+, while Sume Music 1.0 is an API you call from your own code: send a prompt of 1 to 5000 characters to POST /v1/music-1.0/generate and pay a fixed $0.125 per accepted generation. If you want a song inside your own product, ad pipeline or video workflow, the API is the one you can script.

Snap's newsroom post (read 2026-10-04) lists Create Song, which turns a chat into a song, alongside AI Photoshoot, AI Remix, Sticker Remix and AI Fonts. It says the features roll out where Lens+ is available, and Lens+ is a tier within Snapchat+. The page gives no price for Lens+ and no per-song limits.

How do the two compare?

They answer different needs, so the table lists what each source states and nothing more. Where Snap's page is silent, the cell says so.

Create Song and Sume Music 1.0 side by side, read 2026-10-04
QuestionSnapchat Create SongSume Music 1.0
Who can use itLens+ subscribers, where rolled outAnyone with a Sume API key
InputA chatText prompt, 1 to 5000 characters, optional image_url
PriceNot stated on the page$0.125 per accepted generation
Length controlNot stated on the pageNo duration field; structure through cues like [0:00-0:30] Intro
OutputInside SnapchatAudio artifact on Sume media servers, typically audio/mpeg

What does a Music 1.0 call look like?

The request takes a prompt and an optional mode of async, sync, subscribe or webhook. Put exclusions in the positive prompt: a non-empty negative_prompt is rejected with HTTP 400, and so is a duration field. The snippet below uses sync mode and prints whatever the job returns; read the audio from result.artifacts where the type is audio.

import os, requests

resp = requests.post(
    "https://api.sume.com/v1/music-1.0/generate",
    headers={"Authorization": f"Bearer {os.environ['SUME_API_KEY']}"},
    json={
        "prompt": (
            "Upbeat indie pop, 100 BPM. [0:00-0:10] Intro with claps. "
            "[0:10-0:40] Verse and chorus about a Friday night."
        ),
        "mode": "sync",
        "wait_timeout_seconds": 30,
    },
    timeout=60,
)
resp.raise_for_status()
print(resp.json())

Is a text-to-song API the better fit for a video workflow?

When the song has to land inside a video, yes, because you can chain it. Generate the track, place it as the audio spine of a timeline render, and add captions from the lyrics. The existing posts on a lyric video maker and captions from a script cover those steps.

Keep one caveat in mind: the Music 1.0 page says the model is retiring gradually and routes through the Music Router, sume/music-auto. Check the docs page for the current routing before you hard-code anything in production.

What should I verify before shipping a song?

Snap's page does not describe where Create Song output can be posted or reused, so do not assume it leaves Snapchat. For Sume output, treat it as any other generated audio: confirm the destination platform's music rules before you attach it to an ad. The post on adding background music to a video shows the mixing step.

Sources

Related posts

More in Comparisons

All Comparisons posts

Written by Sume