Artlist's Lyria 3 Pro tracks: the same engine through Sume's router
Artlist added Google's Lyria 3 and Lyria 3 Pro for AI music. Sume's Music Router exposes the same engine ids over an API. Compare the two for video beds.

Is Artlist's AI music the same as Lyria 3 Pro? Per Music Business Worldwide (read 2026-10-04), Artlist added Google's Lyria 3 Pro the day after Google announced it, and its AI Music category, built on Lyria 3 and Lyria 3 Pro, launched on March 26, 2026 according to coverage in the search results. It generates from text or up to 10 images and can make tracks up to three minutes long, with a commercial-use license included in Artlist's offering.
Sume's Music Router names the same two Google engines as routable ids, lyria-3.5 and lyria-3-pro, and sends requests with no model to sume/music-auto, which resolves to Lyria 3.5 today. The difference is the surface: Artlist is a creator app with a licence; Sume is an API that returns a file inside a job.
What Artlist reports
The MBW report and Artlist's own prompting guide are the sources for this post. I treat Artlist's licence terms as Artlist's alone; Sume's docs do not describe them and neither does this page.
Side by side
| Question | Artlist AI Music | Sume Music Router |
|---|---|---|
| Engines | Lyria 3 and Lyria 3 Pro | lyria-3.5, lyria-3-pro, or sume/music-auto |
| Where you work | Artlist's app | POST /v1/music-router/generate |
| Image input | Up to 10 images, as reported | One optional public HTTPS image_url |
| Length | Up to 3 minutes for Lyria 3 Pro, as reported | Steer in the prompt; duration is rejected |
| Output | Track inside Artlist | Audio artifact in result.artifacts[] |
Pin the engine
Leave model off to let the router choose, or pass a routable id to pin one. Unknown ids fail with 400 model_not_found and a catalog_url, so a typo does not silently fall back.
import os
import requests
r = requests.post(
"https://api.sume.com/v1/music-router/generate",
headers={
"Authorization": "Bearer " + os.environ["SUME_API_KEY"],
"Idempotency-Key": "artlist-pro-001",
},
json={
"model": "lyria-3-pro",
"prompt": "[0:00-0:30] Intro: sparse piano. [0:30-1:30] Verse: add soft drums. Instrumental, no vocals.",
"mode": "async",
},
timeout=60,
)
print(r.status_code)
print(r.json())Which to use
Pick Artlist if you want its library, apps and licence in one subscription. Pick the Music Router if the track must come back as a file inside an automated pipeline: generate, cut, caption and render in one place. Either way, check the licence that applies to your use before you publish.
Sources
Related posts
More in Comparisons
- Audit a rival price-per-second table against the vendor pages
A Higgsfield post lists Veo 3.1 at $0.40/s and Runway Gen-4.5 at $0.12/s. Google matches the first; Runway plan math gives $0.096 to $0.23. Check yours.
- Background remover API prices: 5 options compared
fal's list prices for five background removers, from $0.0008 per compute second to $0.018 per image, and how Sume's rmbg_create tool compares.
- Claude batch results wait for all; Sume webhooks fire per video
Anthropic Message Batches give results once every request has finished (or after 24 hours). Sume can notify per finished run. Which fits a holiday launch.
- Claude routines vs Desktop tasks vs /loop for recurring Sume renders
Pick between a Claude cloud routine, a Desktop task and /loop for scheduled Sume generation, or a Sume cron schedule when the work is only generation.
Written by Sume