YouTube Shorts Remix is free: what to use for 200 Omni clips a week

Shorts Remix with Gemini Omni is free for creators 18 and over, inside YouTube. For scripted batches, call Omni by API on Sume and mind plan limits.

5 min readSume
All posts

YouTube Shorts Remix with Gemini Omni is free for creators aged 18 and over, but it lives inside YouTube Shorts and the YouTube Create app. The announcement describes it as an in-app feature and does not mention an API, so for a scripted batch of clips you call Omni by API, for example through Sume's /v1/videos with gemini-omni-flash-1.1.

What Remix is

Google's I/O summary says Shorts Remix is free for all users 18 and over, with no paid subscription, and lets people edit Shorts by inserting themselves or visual references. YouTube's own post says the feature carries digital watermarks, identifying metadata and a link back to the original video, and that creators can opt out of visual remixes. Omni is also available at no cost in the YouTube Create app.

Shorts Remix versus an API batch (read 2026-10-05)
NeedShorts RemixOmni by API on Sume
CostNo cost, 18 and overBilled per second, list x 1.25
Where it runsInside YouTubeYour code
VolumeOne person at a timeJobs queue per plan
Source materialOther creators' ShortsYour own clips and references

Know your plan's lane before a batch

Sume admits paid jobs against plan concurrency and a queue. Pro runs 4 jobs at once with a queue of 20, so up to 24 can be accepted; Free runs 1 with a queue of 5. Past that, submit returns 429 queue_full. A full concurrency lane alone is not an error, because jobs wait as queued. Always prefer the generation_limits in the submit response to a static table.

A batch loop that respects 429

The loop submits one job per prompt with its own Idempotency-Key, waits when it sees queue_full, and prints the job ids to poll later.

import os, time, requests

H = {"Authorization": f"Bearer {os.environ['SUME_API_KEY']}"}
prompts = ["A skateboarder lands a kickflip at dawn",
           "A baker pulls a loaf from a stone oven"]
ids = []
for i, p in enumerate(prompts):
    body = {"model": "gemini-omni-flash-1.1", "prompt": p, "duration": 8,
            "resolution": "720p", "aspect_ratio": "9:16"}
    while True:
        r = requests.post("https://api.sume.com/v1/videos", json=body, timeout=60,
                          headers={**H, "Idempotency-Key": f"shorts-batch-{i}"})
        if r.status_code == 429:
            time.sleep(int(r.headers.get("retry-after", 30)))
            continue
        r.raise_for_status()
        ids.append(r.json()["id"])
        break
print(ids)

Choosing between the two

The two can also meet in one process: use the API to produce clean source clips, then let creators remix them inside the app, where the link back to the original is added by YouTube.

  • Pick Remix when a creator wants to react to a specific Short, since that is what it is built for.
  • Pick the API when the same template runs on many rows, such as a product feed.
  • Pick neither for content that needs a licensed person on screen without consent.
  • Re-check the age and region rules on YouTube's page before you plan around Remix.

Weekly cost

At 720p an 8-second Omni clip bills $1.00 on Sume (8 x $0.10 list x 1.25), so 200 clips a week is $200. Remix is free, but it cannot be scheduled, and it is built around remixing existing Shorts. Use Remix for one-off creative play, and the API when the clips come from a spreadsheet. See Generation admission for the queue rules.

Sources

Related posts

More in Use cases

All Use cases posts

Written by Sume