How Base64 Encoding, UTF-8 Byte Mapping & URL-Safe Base64 Work
Why Developers Use Base64 for APIs, Data URIs & Email Attachments
Base64 is a standardized binary-to-text encoding scheme (defined in RFC 4648) that converts arbitrary raw bytes—such as UTF-8 text, images, cryptographic hashes, or binary files—into a safe 64-character ASCII subset (A–Z, a–z, 0–9, +, and /, with = used for padding). Because legacy transport protocols like SMTP email, JSON payloads, XML documents, and HTTP Basic Authentication headers can corrupt raw non-printable bytes, Base64 guarantees data arrives intact across any network. When embedding small icons or fonts directly inside HTML or CSS, developers also use Base64 Data URIs (data:image/png;base64,...) to eliminate extra HTTP requests.
How to Encode or Decode Base64 in 3 Easy Steps
- Select Encode or Decode Mode: Click Encode (Text → Base64) to convert plain text into Base64, or Decode (Base64 → Text) to restore readable UTF-8 text.
- Paste Text or Upload a File: Type any text (including Hindi, Unicode symbols, and emojis) or click + File / Image to Base64 to generate a Data URI locally.
- Copy or Download Output: Toggle URL-Safe Base64 if building JWT tokens or query parameters, then click Copy Output or Download .TXT.
Base64 Encoded Size Expansion Formula & Worked Example
Base64 groups every 3 input bytes (24 bits) into 4 output ASCII characters (6 bits each), increasing payload size by approximately 33.3%:
Encoded Base64 Length (Characters) = 4 × ⌈ Input Bytes / 3 ⌉
Worked Example: Suppose you encode the 3-character ASCII string "Man" (3 bytes: 0x4D 0x61 0x6E).
In binary: 01001101 01100001 01101110 → split into four 6-bit groups: 010011 (19 = T), 010110 (22 = W), 000101 (5 = F), 101110 (46 = u).
Encoded Base64 output = "TWFu" (exact length: 4 × ⌈3 / 3⌉ = 4 characters with zero padding).
Base64 Encoding Standards, Padding Rules & Byte-Size Reference (16 Items)
Use this technical developer cheat sheet to understand Base64 alphabets, padding characters, and size overhead.
| Base64 Concept / Input |
Encoded Output / Rule |
Byte / Bit Structure |
Common Real-World Usage |
| Standard Alphabet (RFC 4648) |
A–Z, a–z, 0–9, +, / |
64 Characters (6-bit) |
MIME email, HTTP Basic Auth, Data URIs |
| URL-Safe Alphabet (Base64url) |
A–Z, a–z, 0–9, -, _ |
Replaces + and / |
JWT tokens, OAuth2 state, URL query params |
| Input Length Divisible by 3 |
No = padding |
3N bytes → 4N chars |
Clean 24-bit block alignment |
| Input Length Remainder 1 |
Ends with == |
1 byte → 2 chars + == |
Pads missing 16 bits to complete 4-char block |
| Input Length Remainder 2 |
Ends with = |
2 bytes → 3 chars + = |
Pads missing 8 bits to complete 4-char block |
ASCII Input: "A" |
QQ== |
1 byte → 4 chars |
Demonstrates double-equals padding |
ASCII Input: "AB" |
QUI= |
2 bytes → 4 chars |
Demonstrates single-equals padding |
ASCII Input: "ABC" |
QUJD |
3 bytes → 4 chars |
Zero padding required |
UTF-8 Rupee Symbol: "₹" |
4oK5 |
3 UTF-8 bytes → 4 chars |
Multi-byte Unicode requires TextEncoder |
| HTTP Basic Auth Header |
Basic dXNlcjpwYXNz |
user:pass → Base64 |
REST API authorization headers |
| PNG Image Data URI |
data:image/png;base64,... |
+33.3% Size Overhead |
Inline small icons & favicons in HTML/CSS |
| SVG Vector Data URI |
data:image/svg+xml;base64,... |
+33.3% Size Overhead |
CSS background-image inline embedding |
| 1 KB Binary File |
~1,368 Base64 chars |
1.33x File Size |
Negligible overhead for tiny assets |
| 100 KB Image File |
~136.5 KB Base64 text |
+33.3 KB Overhead |
Better served as external cached file |
| Is Base64 Encryption? |
No (Encoding Only) |
Zero Secret Key |
Anyone can decode Base64; use AES/HTTPS for secrets |
Native JS btoa() Trap |
Fails on Unicode > U+00FF |
Solved via UTF-8 Bytes |
Our tool safely encodes Hindi & emojis |
Formatting JSON payloads, generating strong passwords, counting characters, or creating QR codes? Explore our companion developer tools: