— 01
JPG with white background
Client exports from Canva, sends a JPG, now there's a white rectangle around their mark. Drop it in, get the logo on transparent. Holds up on serifs and tight letter spacing.
Logo background remover
Clean up
Clean up
any logo file
A JPG with a white background, a screenshot with the logo embedded, a scan from a printed brand book. Drop it in, get a transparent PNG. Holds up on thin strokes, tight kerning, and small marks.
— What this does
Inkurs removes the background from any raster logo file and returns a transparent PNG with a proper alpha channel. Works on JPGs, PNGs with solid-color fills, screenshots, scanned brand books, and logos embedded in PDFs. Built to preserve fine type and thin strokes.
Drop image or paste from clipboard
PNG · JPG · WEBP · HEIC · BMP — up to 20MB
Tip — paste with ⌘+V from anywhere
01 / 01 / Common scenarios
Four ways
logos arrive broken.
Every designer running client work has had this conversation. 'Can you send me the logo?' [JPG attached.] Inkurs is the fastest path from that JPG to something usable.
— 02
Logo on a brand color fill
Placed against solid red, blue, or brand color in a marketing PDF. Inkurs treats any uniform fill as background.
— 03
Screenshot from a website
Logo in a header, you snipped a screenshot. Inkurs isolates just the logo, removes UI chrome.
— 04
Scan from a printed brand book
Scanned page of a brand manual. Paper texture, slight cream tint. Inkurs reads paper as background.
— Worth saying
Raster only.
Inkurs returns a raster PNG. It does not reconstruct a vector file from a JPG. For most working uses a high-resolution PNG is enough; for print at scale, ask the client for the vector source.
02 / 02 / Craft details
Details
that matter for type.
— 01
Thin-stroke preservation
Hairline strokes in serif logos don't get sanded down. Segmentation runs at higher resolution to hold sub-pixel detail.
— 02
Tight kerning isn't eaten
Wordmarks with closed counters (Os, As, Bs) keep negative space. Model doesn't collapse letter interiors.
— 03
Anti-aliased edges
Alpha channel preserves soft anti-aliasing around glyphs. Edges look smooth on any background.
— 04
Small marks stay sharp
Trademark, registered, copyright symbols survive. So do dots, tittles, and accents.
— Frequently asked
Common questions.
Will inkurs return a vector file from my JPG logo?
No. Inkurs returns a high-resolution transparent PNG, not a vector file. A JPG doesn't contain the vector data needed to rebuild SVG or AI files. For a true vector, ask the client for the original design source.
Will it work on logos with thin strokes and tight kerning?
Yes — inkurs runs at higher resolution than typical photo segmentation models to preserve hairline strokes and avoid collapsing closed letter counters (the interior space of O, A, B, etc.).
Can it isolate a logo from a photo, not just remove a flat background?
Not on this tool — that's a different problem called object inpainting. It's on our roadmap. For now, inkurs handles backgrounds (solid, gradient, scene); logo-from-photo extraction needs manual work in Photoshop.
Is inkurs really free with no signup?
Inkurs is free for single-image use with no account, no email, no watermark, and no daily limit. Pro at $3/month adds bulk uploads and API access. The single-image tool stays free, forever.
Do you store or train on my uploaded images?
Inkurs deletes both the uploaded file and the processed output from its servers within 60 seconds of your download. We do not train AI models on user images, share data with advertisers, or retain copies. Privacy policy →
What image formats and sizes does inkurs accept?
Inkurs accepts PNG, JPG, WEBP, HEIC, and BMP files up to 20MB on the free tier. Output is a transparent PNG by default; WEBP is also available. Pro extends the file size cap and adds 8K output.
How does inkurs compare to removebg and Photoroom?
Inkurs gives full HD output free with no signup, while removebg gates HD behind a paid account ($9.99/mo) and Photoroom requires signup for most features ($12.99/mo). Inkurs Pro is $3/month flat. Detailed comparison →