Choose dimensions and a background. The default creates a transparent 1600 × 1200 PNG.
Select a version to preview it in the editor. Closing brings your current text back.
Choose what to restore. This never changes your document text, drafts, or recent files.
Recovery drafts are unsaved copies kept on this device. Restore one to open it, or delete drafts you no longer need.
No recovery drafts are saved on this device.
Choose which button groups to show on the toolbar.
Header/footer need at least 0.4" top and bottom margins
Jump to a line like 42, or a line and column like 42:7.
Image Editor is a free app for creating, opening, and editing images and photos on your computer and in Google Drive.
Open a photo from your computer or Google Drive, edit it with professional image tools, and save it back where it came from or download a new copy.
To get started, open an image or create a new blank image with one of the buttons below.
You're here because you opened an image attachment from Gmail. Image Editor will open it for editing, then let you save a copy to Google Drive or your computer.
To continue, authorize Image Editor with the button below.
You're here to open an image from Google Drive. Image Editor will load it for editing and can save the result back to Drive when you have permission.
To continue, authorize Image Editor with the button below.
You've chosen to create a new image. Choose its size and background while the editor finishes loading.
Select an image with the file picker. In browsers that support direct file access, Image Editor can ask permission to save an edited PNG, JPEG, or WebP back to the same file. Other browsers use a normal download. Either way, decoding and editing stay in the browser tab. Supported inputs include JPEG, PNG, WebP, BMP, GIF, HEIC and HEIF, TIFF, and carefully validated SVG.
Choose Google Drive only when you want it. After you authorize the app, the Drive picker lets you select a supported image. Image Editor requests access to files you choose, not a broad view of everything in Drive. If an image changes elsewhere before you save, the app checks for that conflict and lets you decide whether to keep your edit, use the newer Drive version, or save a separate copy.
Create New Image opens a size dialog instead of guessing. The default is a 1600 by 1200 pixel transparent PNG, with common presets, custom width and height, an orientation swap, and background choices. Large dimensions need much more working memory than the final compressed file, so Image Editor warns before creating an unusually large canvas.
Pintura is the dedicated image-editing component inside the familiar Image Editor shell. The shell owns opening, filenames, saving, Drive, recovery, menus, dialogs, and errors. Pintura owns the pixels and editing gestures, which keeps each responsibility clear and makes Undo and Redo dependable.
Crop removes unwanted edges and can constrain the result to a chosen aspect ratio. Rotation and flipping correct camera angle or composition. Resize changes the actual pixel dimensions of the exported image; it is different from zoom, which only changes how large the image looks while you work. Repeated enlargement cannot recreate missing detail, so start from the best original available.
Filters provide a quick overall look. Fine-tune controls let you make more deliberate adjustments to properties such as brightness, contrast, saturation, exposure, warmth, and clarity. These operations are represented in editable state while the document remains open, so you can revisit decisions, undo them, or return to the home state.
Annotation tools add text, shapes, arrows, lines, and drawings. Decoration and frame tools can give an image a finished edge, while Fill changes the background or surrounding canvas. Image Editor uses system and self-hosted fonts and does not fetch remote stickers or image-backed decoration assets. That keeps a recovery draft self-contained.
Use the Redact tool when visible information must be obscured. Always inspect the final exported file before sharing it. Image Editor rasterizes the result and strips common source metadata, but redaction is only as good as the area you cover. A crop may also be appropriate when sensitive material can be removed entirely.
Digital images are grids of colored pixels. Width and height describe the number of pixels, not a physical paper size. A 1600 by 1200 image contains 1.92 million pixels. During editing, the browser may hold several uncompressed pixel buffers, so memory use can be many times larger than the JPEG or PNG file on disk.
For web pages and email, export near the largest size at which the image will actually be displayed. A smaller image is faster to open and share. For print, keep enough pixels for the intended physical size; printing a tiny source very large reveals softness and artifacts. Cropping first and resizing once usually gives a cleaner result than repeatedly resizing.
PNG and WebP can preserve transparent pixels. JPEG cannot. When a transparent image is exported as JPEG, Image Editor asks for or applies a matte color, white by default, so transparent regions do not turn into an accidental dark background. Choose PNG when crisp transparency matters, such as logos, interface art, or cutouts.
Zooming and panning are workspace controls. They help with precise edits but do not by themselves change output dimensions or quality. The status bar reports image size and zoom separately so it is clear whether you are changing the document or merely the view.
Save is always initiated by you. A writable PNG, JPEG, or WebP opened from your computer or Drive can be overwritten after permission and conflict checks. Converted sources such as HEIC or TIFF begin with Save As because writing them back would silently discard container features. Once you save to a supported home, later Save commands can update that new file.
Save updates the current home. Save As writes a new file and adopts it as the document home only after the write succeeds. Download a Copy creates an export without changing the home or pretending that newer edits are safely stored. This distinction matters: a download should never clear recovery or mark the working document clean by accident.
PNG is lossless and is the safest choice for transparency, text, diagrams, and repeated editing, though photographs can be large. JPEG is widely compatible and compact for photographs, using a high-quality setting of 0.92. WebP often produces a smaller photo file at quality 0.80 and can retain transparency. Image Editor keeps the filename extension and MIME type together, so a file named .png is actually encoded as PNG.
The exported image is newly rendered. Common EXIF, GPS, IPTC, XMP, ICC, comment, and container metadata from the source are not copied. Supported camera orientation is baked into the pixels so the result displays consistently. This improves privacy, but you should still examine the visible image for addresses, documents, reflections, or other information you did not intend to share.
JPEG, PNG, and WebP are browser-friendly editing and output formats. Other inputs are normalized to a PNG working image before Pintura receives them. Image Editor retains both the untouched original and the exact normalized asset for recovery, but the export intentionally does not reproduce every feature of the source container.
HEIC-family images are decoded locally with the repaired libheif 1.23.1 and libde265 1.1.1 subsystem. The primary image becomes a PNG working asset. Image sequences, HDR behavior, depth data, and other container features are not preserved. The decoder runs in a one-shot worker with source, pixel, item, tile, frame, time, and memory limits; very large or malformed files are refused.
For TIFF, Image Editor selects the largest supported RGB page or sub-image and records its orientation; multipage structure and high-bit-depth data are not preserved. Animated GIF opens its first frame with a notice. BMP becomes PNG. SVG is accepted only after strict validation, then rasterized to PNG. Scripts, external resources, document types, entities, event handlers, foreign objects, imports, compressed SVGZ, and pathological dimensions are rejected.
Overwriting a GIF with a still frame, a multipage TIFF with one page, or a vector SVG with a raster image would be surprising and destructive. Image Editor therefore routes converted inputs to Save As and offers PNG, JPEG, or WebP. The conversion notice explains what will be lost before you choose an output.
Editing state and durable saving are separate. After you pause, Image Editor stores a recovery checkpoint in browser storage when available. “Recovery updated” does not mean “Saved”: it only means the browser has a local draft that may help restore this tab after a crash or accidental close.
The source asset is stored once, along with the normalized asset when conversion was required. Later checkpoints store canonical editor state and export choices rather than another full image. Recovery starts a fresh Undo baseline because the complete transient history stack is not persisted. A storage-health banner tells the truth if quota, private browsing, or another browser restriction prevents reliable recovery.
The start page can list up to ten reusable Drive files or local file handles. These entries stay on your device. Tab identity, browser locks, and draft ownership reduce the chance that two tabs silently claim the same recovery record. Manage Drafts lets you inspect abandoned work, while live dirty drafts are not deleted merely to meet a count.
Rendering and writing are serialized. Each save snapshots the document generation, edit sequence, home, conflict baseline, image state, and output settings, then rechecks them after asynchronous steps. If you keep editing while an export is rendered, the completed write covers the earlier snapshot and the newer edit remains visibly dirty and recoverable.
Image pixels, image state, annotations, filenames, paths, Drive IDs, metadata, and decoder errors are not sent to Visware, Pintura, analytics, or advertising services. The editor libraries, decoder payload, fonts, and shell icons are served with the app rather than fetched from a third-party CDN. Google receives a file only when you explicitly use Google Drive, and your browser contacts advertising or analytics services only under the site’s configured consent posture.
Operational events use stable categories such as source kind, image format, coarse byte and pixel ranges, duration ranges, result codes, and chosen editing utility. They exclude filenames, URLs, query strings, raw exceptions, and image content. Launch URLs are scrubbed before analytics or ads initialize, and the reported page location contains only the canonical origin and path.
Local processing avoids a server upload, but it does not make memory unlimited. A compressed camera photo can expand into several large RGBA buffers during decode, editing, and PNG creation. Image Editor allows one active decoder, cancels stale work, warns about large canvases, and enforces hard decoder limits. Closing other memory-heavy tabs can help with a legitimate large image.
Image Editor is free to use. No account is needed for images on your computer. Google authorization is requested only for Drive actions, and short-lived access tokens stay in memory rather than being stored as application data.
No. Saving is a user action. Eligible files can be written back only after the browser grants permission and conflict checks pass. Converted formats use Save As. You can always choose Download a Copy to leave the current home unchanged.
Read the full FAQ, visit the help center, review the privacy policy, or contact support@imageeditor.app. The seven links in the footer also cover installation, terms, product background, and support.
We're sorry - Image Editor was unable to load.
Please see the troubleshooting options below.
Still not working?
If none of these steps resolve the issue, please email support@imageeditor.app with the error message below so we can assist you directly.