Remove the exact distraction you mask, reconstruct what was behind it, and keep transparent assets composable with native RGBA cleanup.
Drag to compare
Pick the workflow that looks most like your backlog, then use the same model through Jasper Image API, Jasper-managed image pipelines, or sales-led solution design.
A smaller set of proof examples, chosen to show the cases a buyer or technical evaluator is most likely to care about.
Enough technical detail to qualify the endpoint quickly, with the full reference one click away.
| Endpoint | POST /v1/image/cleanup |
| image_file | Source image · Image to clean |
| mask_file | Region to remove · Binary B&W, exact same dimensions as source |
| Accept (header) | Output format · Use image/png to preserve alpha · Default: varies |
Input prep matters: use a clean, pixel-matched binary mask with enough margin around the object so the model can remove the full target.
curl -X POST https://api.jasper.ai/v1/image/cleanup \
-H "X-API-Key: $JASPER_API_KEY" \
-H "Accept: image/png" \
-F "[email protected]" \
-F "[email protected]" \
-o cleaned.pngimport os, requests
resp = requests.post(
"https://api.jasper.ai/v1/image/cleanup",
headers={"X-API-Key": os.environ["JASPER_API_KEY"],
"Accept": "image/png"},
files={"image_file": open("photo.png", "rb"),
"mask_file": open("mask.png", "rb")},
timeout=120)
open("cleaned.png", "wb").write(resp.content)The full Cleanup benchmark helps teams review mask-driven object removal, alpha preservation, reconstruction quality, and hard cases that need human review.
Request access when you want to compare transparent PNG behavior, ghosting risk, border artifacts, or category-specific Cleanup cases before production rollout.
See visual cases, methodology, output notes, and the model behavior that matters for your category.
Access benchmarkAI image cleanup removes a selected object, person, defect, or distraction and reconstructs the area behind it. Jasper Cleanup gives teams precise, mask-driven object removal for production assets and automated image workflows.
Jasper Cleanup combines precise mask control with context-aware reconstruction and native RGBA support. That gives teams clearer removal boundaries and keeps transparent assets composable instead of flattening them onto an opaque background.
Yes. Jasper Cleanup uses a black-and-white mask that matches the source image dimensions, which gives users exact control over what is removed. If your product has no mask editor, Jasper can demonstrate a sales-led agentic workflow that creates and handles the mask for you.
Jasper Cleanup can remove product props, dust, defects, unwanted people, signs, shadows, reflections, and other visible distractions when the target is covered by a mask.
Yes. Jasper Cleanup supports native RGBA object removal and can preserve the alpha channel of transparent PNG assets. Use PNG output when the cleaned image needs to remain transparent and composable.
Mask the entire target with a small margin, including connected shadows or reflections. A clean, pixel-matched mask gives Jasper enough context to remove the full object and reconstruct the background with fewer visible remnants.
In Jasper's June 2026 benchmark of 14 images across six providers, Jasper Cleanup was strongest overall, with cleaner erases, preserved alpha, and no rejected inputs. The comparison is anonymized publicly, and teams can request benchmark access to review the relevant cases.
No image cleanup model is perfect. Jasper Cleanup performed strongest overall in the June 2026 benchmark, but difficult scenes can still leave a faint trace, so production workflows should review uncertain outputs before publishing.
Bring your image set, production constraints, or benchmark questions. Jasper can help map the right API, app, agent, or pipeline path.
