remove.bg API Alternative — Background Removal on KeyoAPI

If you integrated remove.bg’s self-serve API and need a prepaid matting endpoint that stays under your control after Dec 1, 2026, KeyoAPI exposes BRIA RMBG-2.0 on POST /v1/images/mattings — same Keyo key you already use for chat, Whisper, and TTS. This page is for developers migrating API traffic, not a free browser cutout tool.

remove.bg API Alternative is a model-family pricing lane on KeyoAPI — a prepaid AI API gateway where one Bearer key and one wallet ($1 = 500,000 quota units) cover chat, speech, vision, OCR and video models.

Migration options at a glance

Indicative figures for planning. Confirm live Keyo sell rates on interactive Model Square or the static pricing list.

Independent API option

KeyoAPI is not Canva, not Leonardo.Ai, and not a claim to be remove.bg’s official successor. Consumer click-to-cutout traffic is pointed at Canva by remove.bg itself. Keyo’s lane is developer-metered matting on a multi-model relay: authenticate, upload, get alpha — then continue to OCR, TTS, or chat on the same balance.

Migration details

Published transition (verify on remove.bg FAQ / API pages): standalone remove.bg website unavailable from 1 December 2026, 9:00am CET; self-serve API background removal moves to Leonardo.Ai; existing Enterprise API contracts are described as unaffected. Unused PAYG credits expire at that cutoff — treat vendor pages as source of truth if wording changes.

Keyo sell rate for RMBG-2.0 is about $0.0068 per request (indicative). Confirm live at the RMBG-2.0 model page. Typical legacy SaaS remove-bg APIs often sit near $0.05–$0.20 per image on public marketing pages; use that only as planning context, not a contractual quote. Worked example: 10,000 product photos / month at ~$0.0068 ≈ $68 indicative on Keyo RMBG — recompute with your live rate and retry policy.

Migration checklist for API traffic: (1) inventory every remove.bg call site and note image size / QPS; (2) create a Keyo key at the sign-up page and top up Wallet; (3) open the RMBG-2.0 model page for interactive try/buy (no separate upload playground); (4) switch clients to POST multipart https://www.keyoapi.xyz/v1/images/mattings with model=RMBG-2.0 and field image=@file; (5) compare alpha quality on a golden set of SKUs before cutting over; (6) keep a feature flag so you can roll back during the Dec 2026 window.

When plain subject/background cutouts are not enough — for example open-vocabulary masks, multi-object labels, or promptable regions — escalate to sam3 via POST /v1/images/segmentation. Guides: the RMBG-2.0 model page and the sam3 model page. Price row also appears on the model comparison page.

Pipeline tip: many shops matte first, then run OCR (see the Unlimited-OCR model page or the MinerU2.5-Pro model page), then chat or TTS on the same prepaid key. If the next step is speech or talking video, continue to the TTS API guide or the AI avatar video generator page without opening a second vendor wallet.

Operational notes: send reasonable image sizes (downscale huge originals before upload), log HTTP status and Keyo request ids for support, and treat matting as a metered request — empty or corrupt files still consume budget when accepted. Confirm live rates after every catalog change on Model Square.

# cURL — RMBG-2.0 matting (multipart)
curl https://www.keyoapi.xyz/v1/images/mattings \
  -H "Authorization: Bearer $KEYO_API_KEY" \
  -F model=RMBG-2.0 \
  -F image=@./product.png \
  -o ./product-alpha.png

# Python — requests multipart
import requests
url = "https://www.keyoapi.xyz/v1/images/mattings"
headers = {"Authorization": "Bearer YOUR_KEYO_API_KEY"}
files = {"image": open("product.png", "rb")}
data = {"model": "RMBG-2.0"}
r = requests.post(url, headers=headers, files=files, data=data, timeout=120)
r.raise_for_status()
open("product-alpha.png", "wb").write(r.content)

Get started in five steps

1. Create an account — new users receive $10 of gift credit for prototyping.
2. Create an API key in the console.
3. Point your OpenAI SDK at https://www.keyoapi.xyz/v1 with the model ID from this page, or call the dedicated REST path from the API handbook.
4. Track spend in the console Wallet — $1 equals 500,000 quota units, and gift credit burns before recharged balance.
5. Set a per-key spending cap before wiring production traffic.

A verified first call is the acceptance test for the setup: the response should carry the payload shape documented in the handbook, the Wallet ledger should show a deduction in quota units (500,000 units = $1), and repeated calls should deduct linearly at the listed rate. Any mismatch usually means the model ID or endpoint is wrong — check both against the pricing list before debugging client code.

FAQ

Also see the model comparison page, the gift-credit models page, the full pricing list, and RMBG-2.0 · sam3.