Is Z-Image Turbo on Sume? Not listed: the cheapest rows instead

Answer

No. Sume's image catalog on 2026-10-04 lists 19 rows and none of them is Z-Image or Z-Image Turbo. You can confirm it with GET /v1/images/models and search the ids; if a row is added later, it will appear there first.

If you want a very cheap image, the listed alternatives are Higgsfield Soul at $0.005 per image (text-only), then Grok Imagine and Qwen Image at $0.025, both with reference editing.

Check the catalog

This script lists every model id and name that contains a keyword, so you can test for Z-Image, or any other family, without reading the page.

import asyncio, os, httpx

async def main():
    async with httpx.AsyncClient(timeout=30) as client:
        r = await client.get(
            "https://api.sume.com/v1/images/models",
            headers={"Authorization": f"Bearer {os.environ['SUME_API_KEY']}"},
        )
    rows = r.json()["data"]
    hits = [m["id"] for m in rows if "z-image" in (m["id"] + m["name"]).lower()]
    print(len(rows), "rows;", hits or "no z-image row")

asyncio.run(main())

What to do

• Use sume/auto if you want Sume's image router to pick; it does not disclose which family ran.
• Pick the cheapest row that has the features you need: Qwen Image for batched edits, Imagen 4 Fast for text-only batches of four.
• Do not assume an unlisted model is reachable under an alias. Only the ids in the catalog, plus the documented legacy Image Router ids, are accepted.

Prices are Sume billable amounts, provider list price times 1.25. Docs: Sume Image API.

Before a large run

Prices and descriptors change when the catalog changes, so confirm them before you spend. Call GET /v1/images/models/{id}/endpoints for the row you plan to use and read its pricing line and supported_parameters; both come back in one response.

Then run a pilot of three to five images and read usage.cost on each response. Multiply by your planned count for a forecast you can trust. Completed generations are billed in full and failed or cancelled ones are not, so a pilot that errors costs nothing.

For big batches, use mode: "async" or mode: "webhook" with a public HTTPS webhook_url, so no request waits on the 30-second sync limit. Poll GET /v1/jobs/{id}/status and fetch GET /v1/jobs/{id}/result when the job completes.