A small fingerprint.
A precise comparison.

Generate SHA-256, SHA-384, SHA-512, SHA-224, SHA-1 and RIPEMD-160 checksums for text or files. Select one or several algorithms. SHA-256 is selected initially. SHA-1 and RIPEMD-160 are included for legacy compatibility, not as recommendations for new security designs. MD5, HMAC, password hashing, digital signatures and file scanning are not provided.

Paste a known expected checksum, select its algorithm and choose Hex or standard Base64. The tool compares it with each calculated result. Hex is case-insensitive; Base64 is case-sensitive and must be canonical. Incorrect lengths, filenames pasted with digests and invalid characters are reported rather than silently removed. The chosen algorithm must have been calculated. An empty expected value means no verification was requested.

A match means the calculated checksum equals the supplied value. It does not prove that a file is harmless, authentic or from a trustworthy publisher. Obtain the expected value from a trusted source independently of a questionable download. Two files with matching hashes are not a mathematical proof that their contents must be identical. This is a checksum utility, not malware detection, encryption or a password-storage system.

Text mode hashes the UTF-8 bytes of the editor contents, with no trimming, Unicode normalization or automatic newline added. The browser textarea may normalize line endings during editing. Hex and Base64 modes hash decoded bytes, not the encoded characters. File mode hashes the original bytes, including BOM, line endings, hidden data and image metadata. A filename does not contribute to its content hash. Use file mode when exact original bytes matter.

UTF-8 text, hexadecimal byte pairs, standard Base64 and URL-safe Base64. ASCII spaces, tabs and line breaks are ignored in hex and encoded Base64 input. Hex does not accept 0x prefixes or incomplete byte pairs. Standard Base64 requires correct padding; URL-safe Base64 allows omitted padding. Invalid alphabets, invalid lengths and nonzero unused bits are rejected. Empty text or an empty file is valid and has a defined checksum.

Text or decoded byte input is limited to 1 MiB, with at most 3 MiB of encoded characters. Select up to 10 files, each up to 32 MiB, and up to 64 MiB total. Files are processed sequentially in 64 KiB chunks in a separate worker. Progress counts input bytes; generating several algorithms takes longer. A run has a 5-second startup deadline and 60-second processing deadline. Cancel stops it, and failures return no partial report.

Copy an individual digest in lowercase hex, uppercase hex or Base64. Copy or download a JSON report with all selected algorithms, byte counts, filenames and verification results. The TXT report uses the selected display format and quotes filenames to preserve special characters. The TXT report labels its digest encoding. File batches can also export a GNU-style SHA checksum file for a calculated SHA algorithm, provided filenames are unique. Use Compare beside any digest to verify other batch members against it. Reports exclude the original text and file contents. Display or comparison changes refresh reports without rehashing completed input.

The selected text and files are processed locally in your browser. The tool does not upload file contents, fetch remote URLs or save input history. Clear removes inputs and temporary reports. Normal site analytics and feedback services may still load. Downloaded reports contain filenames and checksums, so review them before sharing.