Best for · Any job with one image or a repeatable batch
Build a batch in the workspace
How to use it
- 1
Choose Compress, Resize, Crop, Convert, Generate pixel art, Add watermark, AI watermark removal, AI background removal, or AI upscaling from the operation rail.
- 2
Select Add images for specific files, Add folders to discover supported files inside a folder, or drag files directly into the window.
- 3
Wait until each row has finished reading its format, dimensions, size, and status. Options stay locked while files are still being prepared.
- 4
Remove unwanted rows or use Clear to start over. Reset options restores the current tool's defaults without removing files.
Controls explained
- File rows
- Show the original format, dimensions, file size, processing state, and the processed result when one exists.
- Thumbnail
- Opens or locates the source image; thumbnail generation does not block metadata loading or processing.
- Cancel processing
- Stops the active batch. Files that have not completed are restored to their previous usable state.
- Demo image
- Loads a safe sample when you want to understand a tool before using your own files.
Before you process
- PDF files are accepted only by Compress. Switching to another operation removes PDFs from the current list.
- A full trial is available before purchase so you can evaluate the complete workflow with your own files.
A ready file list with one shared set of operation options that can be applied consistently across the batch.
Best for · Web delivery, email attachments, storage reduction, and lighter documents
Compress images and PDFs
How to use it
- 1
Choose Compress and add supported images or, with an activated license, PDF files.
- 2
Review the format groups shown in the options panel. A mixed batch can expose several groups at once.
- 3
Set clarity, colors, lossless mode, or compression type per format group, or apply one of the built-in presets. Start near the defaults, then lower clarity gradually when size matters more.
- 4
Run Compress. Compare the original and processed size in each row, preview visual changes where available, and save the accepted results.
Controls explained
- Presets
- Web, Social, Email, and Archive set sensible values for every format group at once; changing any option switches to Custom.
- Metadata
- Keep all metadata, keep only the colour profile, or strip everything for the smallest, most private file.
- Target size
- For JPEG, WebP, AVIF, HEIC, and HEIF, set a maximum file size in KB and PicSmaller searches for a quality that fits.
- JPEG clarity and colour detail
- Balances visual detail and file size; the colour-detail option keeps full colour sampling for red text, charts, and illustrations at a larger file size.
- PNG lossless, clarity, and colors
- Enable lossless mode to repack the file without changing a single pixel, or tune clarity and palette size for smaller lossy results.
- GIF colors and strength
- Control palette size and lossy reduction while preserving animation timing and looping; lossless mode keeps every original color.
- WebP and AVIF
- Choose quality for lossy output or enable lossless mode.
- HEIC / HEIF and JXL
- Use the format-specific clarity control to balance detail and size.
- TIFF
- Choose Deflate, LZW, or JPEG compression. JPEG quality appears only for JPEG-compressed TIFF output.
- PDF image clarity
- Recompresses eligible embedded images while preserving text, vectors, page dimensions, and document structure.
Before you process
- Already optimized or mostly text/vector PDFs may not become smaller.
- Image preview is not available for PDF compression.
- Presets are a convenient starting point; fine-tune any format group afterwards for full control.
A same-format processed file. When the processed version is not smaller, PicSmaller reports that the original was kept.
Best for · Store listings, social posts, email assets, thumbnails, and consistent exports
Resize an entire image set
How to use it
- 1
Choose Resize and add images.
- 2
Select a resize method: width, height, short edge, long edge, percentage, ratio, or exact width and height.
- 3
Enter integer dimensions from 1 to 100,000. Enable Shrink only when small originals must never be enlarged.
- 4
Run Resize, inspect the processed dimensions in the list, and use Preview to check sharpness and geometry before saving.
Controls explained
- Use this width / height
- Sets one edge and preserves the original aspect ratio.
- Use a percentage
- Scales both dimensions by the same percentage of the original, from 1% to 400%.
- Set the short / long side
- Adapts automatically to portrait and landscape orientation while preserving aspect ratio.
- Use this ratio
- Changes output to the entered width-to-height proportion; a ratio different from the original can stretch the image.
- Exact width and height
- Forces both dimensions. Mismatched proportions can visibly distort the image.
- Exact-size fit
- For exact dimensions, choose Fill to keep the aspect ratio and crop the overflow, Contain to fit the whole image with a background, or Stretch to force both dimensions.
- Resampling
- Auto suits photos, Smooth reduces halos around high-contrast lines, and Sharp keeps clean blocks when enlarging pixel art, icons, and line art.
- Shrink only
- Leaves images below the requested size unchanged instead of enlarging them.
Before you process
- Use Crop rather than exact resize when composition must change without distortion.
- For mixed orientations, long-side or short-side rules usually give the most consistent batch.
Images with dimensions calculated from the selected rule. Animated GIF and WebP playback is preserved.
Best for · Avatars, cover images, product tiles, reels, banners, and uniform catalogs
Crop by size or aspect ratio
How to use it
- 1
Choose Crop and add images.
- 2
Select By width and height for a fixed pixel crop, or By ratio for a proportional crop.
- 3
For ratio crops, choose 1:1, 4:3, 3:2, 16:9, 9:16, 4:5, 3:4, 2:1, or enter a custom width and height ratio.
- 4
Choose Follow subject to keep the most interesting area of each image, or pick one of nine fixed anchors, run Crop, and verify the composition in Preview.
Controls explained
- Crop size
- Defines the requested pixel width and height. If the request exceeds the source, the crop area is reduced, or you can enable padding to keep the requested size with a background border.
- Aspect ratio
- Finds the largest matching rectangle inside each source image.
- Nine-point anchor
- Keeps content toward top-left, top, top-right, left, center, right, bottom-left, bottom, or bottom-right.
- Follow subject
- Analyzes each image and keeps the area most likely to contain the subject, useful when composition varies across a batch.
- Custom ratio
- Accepts integer width and height components from 1 to 100,000.
Before you process
- Center works for balanced compositions; choose Follow subject or a side anchor when subjects are not centered.
- Crop removes pixels and does not generate missing content. Use the padding option to keep a fixed size for smaller images.
A cropped image mapped to the selected source region. Animated GIF and WebP timing and looping are preserved.
Best for · Web compatibility, modern formats, print workflows, archives, and image-to-PDF export
Convert between 16 output formats
How to use it
- 1
Choose Convert, add images, then select JPEG, JPG, PNG, GIF, WebP, AVIF, HEIC, HEIF, TIFF, TIF, BMP, SVG, JP2, J2K, JXL, or PDF.
- 2
Set options shown for the selected destination: clarity, compression level, colors, dithering, lossless mode, TIFF compression, metadata, or background color.
- 3
Read the animation notice before processing. Still formats keep only the first frame of animated input; converting a still image to GIF or WebP does not create animation.
- 4
Run Convert and confirm the new format and size in the list before saving.
Controls explained
- JPEG / JPG
- Adjust output clarity for photographs and broadly compatible web assets.
- PNG
- Choose compression level; higher levels can be smaller but slower.
- GIF
- Set palette colors and dithering for smoother gradients.
- WebP / AVIF
- Adjust quality or use lossless output.
- HEIC / HEIF
- Adjust quality; animated inputs produce a static first-frame result.
- TIFF / TIF
- Choose Deflate, LZW, or JPEG compression and quality where applicable.
- JP2 / J2K and JXL
- Adjust output quality for archival and modern formats.
- Creates one PDF page for each input image. Animated inputs use the first frame.
Before you process
- PDF output requires an activated license; the other listed destination formats are available in the trial.
- Conversion changes the container and encoding; it does not improve a low-quality source by itself.
One processed file per input in the chosen destination format, with explicit warnings for any animation change.
Best for · Game assets, retro portraits, stylized product art, avatars, and cohesive pixel-art sets
Generate pixel art in batches
How to use it
- 1
Choose Generate pixel art and add one image or a folder. Every file in the batch uses the same settings.
- 2
Set Pixel size, Color count, Detail, and Color sampling. Flat blocks keep crisp dominant colors; Blended sampling creates softer transitions.
- 3
Choose automatic colors, a theme color with generated light and dark shades, a classic palette preset, or your own imported palette.
- 4
Open Advanced parameters to choose optional dithering and strength, outline style and width, isolated-pixel cleanup, and alpha threshold. Pick an output format and scale, enable per-frame processing for animations when needed, then generate and preview at 100%.
Controls explained
- Pixel size
- Controls how many source pixels become one logical pixel. Higher values create larger blocks and a lower-detail result.
- Color count and palettes
- Limit the result from 4 to 256 colors. Colors can come from the source automatically, from a theme color's light-to-dark ramp, from classic palette presets, or from an imported GPL, HEX, TXT, PAL, or ACT palette. The current palette can be exported as GPL, HEX, or PAL.
- Color sampling
- Flat blocks use each cell's dominant color for crisp edges. Blended sampling averages the cell for softer transitions and may introduce intermediate colors.
- Detail and color reduction
- Balance smoothing and sharpness. Color reduction is selected automatically for consistent results without an extra algorithm setting.
- Dithering
- Choose None, Soft blend, Clean grain, Natural grain, Retro pattern, Coarse pattern, or Fine pattern. Strength controls how visible the selected texture becomes.
- Outline
- Optional black, dark, or colored outlines trace transparent-subject edges and interior color-block boundaries, and can be one or two logical pixels wide.
- Output scale and format
- Choose source dimensions with hard nearest-neighbor blocks, or export a true pixel grid where every block is one pixel for sprite editors and game engines. Output to PNG, WebP, GIF, or JPEG.
- Cleanup and alpha threshold
- Remove stray isolated pixels and control how transparent a pixel must be before it is kept, preserving soft edges such as hair.
Before you process
- This tool does not use an AI model or require a network connection; processing runs locally on the CPU.
- Animated inputs stay static unless Process all frames is enabled; when enabled with WebP or GIF output, every frame is converted with one shared palette.
- PNG, WebP, and GIF preserve transparency and hard edges. JPEG has no transparency and uses a solid background.
A pixel-art image at the chosen output scale for each input, region-averaged, color-limited, and treated in a consistent batch workflow.
Best for · Branding, proof delivery, ownership labels, and campaign attribution
Add text or image watermarks
How to use it
- 1
Choose Add watermark and select Text watermark or Image watermark.
- 2
For text, enter the wording, system font, and color. For an image mark, choose a local logo or graphic.
- 3
Choose One watermark or Repeat across image. A single mark can be placed at a corner or in the center.
- 4
Tune opacity and size; for tiled marks, adjust spacing. Process and preview the mark on light and dark parts of the image before saving the batch.
Controls explained
- Text style
- Uses an installed system font and a selected text color.
- Image watermark
- Uses a local image file and requires the corresponding licensed feature.
- Opacity
- Ranges from 10% to 90%. Lower values are subtler; higher values are harder to overlook.
- Size
- Ranges from 6% to 24% of each image. Around 10% is suitable for many single marks.
- Tile spacing
- Ranges from 70 to 160 and controls density when repeating the mark.
- Position
- Single marks support top-left, top-right, center, bottom-left, and bottom-right.
Before you process
- Use sufficient contrast for the source set; one color may not remain readable across every image.
- Only watermark images you own or are permitted to modify.
A processed image with the selected mark rendered consistently relative to each image's dimensions.
Best for · Product photos, portraits, catalogs, presentations, and design cutouts
Remove backgrounds with AI
How to use it
- 1
Choose AI background removal and add images with a clearly distinguishable subject.
- 2
Choose Transparent for reusable cutouts or Solid color for a finished background, then select the color if needed.
- 3
Choose Standard for faster batches or High when hair, fine edges, and small details matter more than processing time.
- 4
On first use, approve the required model download. After it is stored locally, processing can run offline. Process and inspect edges in Preview.
Controls explained
- Transparent background
- Outputs a PNG with alpha so the subject can be placed over other designs.
- Solid color
- Composites the subject onto the selected color and still outputs PNG.
- Standard quality
- Uses a smaller, faster model suited to everyday images and batches.
- High quality
- Uses a larger model to preserve more edge detail and takes noticeably longer.
Before you process
- Animated inputs keep only the first frame and become static PNG files.
- Fine transparent objects, motion blur, or subject colors close to the background deserve a careful preview.
A static PNG with the source dimensions, either transparent or filled with the chosen color.
Best for · Cleaning authorized source assets when an unmarked original is unavailable
Remove visible watermarks with AI
How to use it
- 1
Choose AI watermark removal and add authorized images.
- 2
Select Standard for faster batches or Detailed to detect smaller watermark edges.
- 3
Start with Balanced cleanup. Use Conservative to protect nearby detail or Strong when faint edges remain.
- 4
Download the two required model resources on first use, process the images, and inspect every repaired region at a useful zoom before saving.
Controls explained
- Standard / Detailed
- Detailed finds finer boundaries but requires more processing time.
- Conservative
- Limits the repair area to reduce changes to nearby image detail.
- Balanced
- Default compromise between complete cleanup and preserving surroundings.
- Strong
- Expands cleanup when edges remain, with a greater chance of altering nearby detail.
Before you process
- Complex texture, translucent marks, and very large watermarks may need manual review.
- If no visible watermark is detected, PicSmaller keeps the image unchanged.
- Animated input becomes a static first-frame PNG.
A static PNG at the original dimensions with detected regions reconstructed, or the original appearance when no mark is found.
Best for · Small product photos, older assets, presentation graphics, and higher-density displays
Upscale small images with AI
How to use it
- 1
Choose AI image upscaling and add the source images.
- 2
Choose 2× for a moderate size increase or 4× when substantially more output resolution is required.
- 3
Use Standard for general work or Detailed when fine texture matters and extra processing time is acceptable.
- 4
Approve the first-use model download, process, then inspect the result at 100% rather than judging only from a fit-to-window view.
Controls explained
- 2×
- Doubles width and height, producing four times the pixel count.
- 4×
- Quadruples width and height, producing sixteen times the pixel count and requiring more memory and time.
- Standard quality
- Faster choice for routine batches.
- Detailed quality
- Prioritizes fine reconstruction and takes longer.
Before you process
- AI cannot recover text or detail that is completely absent from the source; inspect critical content carefully.
- For processing failures, try a smaller source or the 2× scale.
- Animated input keeps only its first frame.
A higher-resolution static image at 2× or 4× the original width and height.
Best for · Quality control and deliberate output management
Compare results and save safely
How to use it
- 1
After processing, choose Preview on a completed row to open the original and processed result on one aligned canvas.
- 2
Use the shared zoom controls or the mouse wheel. Above the fit level, drag either side to pan both views together.
- 3
Use Save as for one item or Save all as to choose a destination folder for every completed result.
- 4
Choose overwrite only when you intentionally want to replace source files. Confirm the warning and keep a backup for irreplaceable originals.
Controls explained
- Synchronized comparison
- Original and processed views share geometry, zoom, and pan so details remain aligned.
- Fit and zoom
- Fit shows the complete image; zoom supports detailed inspection from 50% to 1600%.
- Save as
- Copies one processed result to a path you choose.
- Save all as
- Writes all completed results to one folder and resolves duplicate names with numeric suffixes.
- Overwrite originals
- Explicitly replaces source targets after confirmation and requires the licensed entitlement.
Before you process
- PDF compression cannot be opened in the image comparison window.
- If destination files already exist, PicSmaller asks for confirmation before replacing them.
Reviewed output stored at a chosen location, while originals remain untouched unless overwrite was explicitly confirmed.
Best for · Multi-step production such as resize → watermark → convert
Chain operations and keep useful settings
How to use it
- 1
Open Settings from the title area.
- 2
Enable Keep working from the last result when the next operation should consume the previous output. Leave it off when every tool should start from the untouched original.
- 3
Enable Save last operation settings to restore the active tool and its parameter values in future sessions.
- 4
Choose a temporary folder with sufficient free space. Save settings, then process and switch tools to build the desired chain.
Controls explained
- Continue from last result
- Feeds the latest processed file into the next operation for the same row. This is a licensed capability.
- Save last operation settings
- Persists the current operation and its controls after short changes and on exit.
- Temporary folder
- Stores previews, thumbnails, intermediate files, AI outputs, and unsaved processed results.
- Language
- Changes application UI language without changing image metadata or output.
Before you process
- When continuous processing is off, switching tools and processing again uses the original source.
- Temporary results are not a substitute for saving completed output to a permanent folder.
A repeatable workspace that either protects the original at every step or intentionally builds a multi-operation pipeline.
Best for · Unlocking licensed tools and managing an activated computer
Activate, work offline, and recover a device
How to use it
- 1
Select the license status in the application and enter the activation code exactly as issued.
- 2
Stay online while activation completes. After success, open the license status again to review expiry or lifetime status and the devices currently in use.
- 3
Continue normal local work as usual. If the application asks for a connection for licensing actions, reconnect and follow the on-screen guidance.
- 4
If you replace a device or need help restoring access, reconnect and select the license status, or contact support for assistance.
Controls explained
- Full trial
- Provides every feature and unlimited batches during the trial period before purchase.
- Activated
- Unlocks the feature, batch, save, device, and duration entitlements defined by the purchased plan.
- Device list
- Shows current and other registered devices with platform, application version, and authorization status.
- Recovery
- Helps you continue using an eligible license after a device change or support action.
Before you process
- Activation, recovery, and some licensing actions require network access.
- Keep your activation code available and follow the in-app status guidance when prompted.
- If you cannot restore access normally, contact support with your order details.
Purchased features remain available for ordinary local work according to your active plan.