Base64 Encoder & Decoder
Encode plain text to Base64 or decode Base64 back into readable text instantly. Great for developers, APIs, JSON payloads, data URLs, and quick debugging.
13 characters, 13 bytes
20 characters, 20 bytes
Data URL and file helper
Base64 often appears inside data URLs like data:image/png;base64,....
Detected or suggested prefix
data:text/plain;base64,
Choose any file and this tool will load it as a full data URL. Switch to Decode to inspect the original text payload, or copy the Base64 string for embeds and APIs.
Base64 tips
- Base64 encodes binary data as text, but it is not encryption.
- URL-safe Base64 swaps + and / for - and _.
- Encoded output is usually about 33% larger than the original bytes.
- You can paste full data URLs directly, and the tool will detect the prefix automatically.
How to Use
- Choose Encode to turn plain text into Base64, or Decode to convert Base64 back into text.
- Paste or type into the input box and the output updates live as you work.
- Enable URL-safe mode when you need Base64 that uses - and _ instead of + and /.
- Use the copy buttons to grab either the input or output quickly.
- Paste a full data URL or upload a file to inspect its Base64 prefix and payload.
Frequently Asked Questions
What is Base64 used for?
Base64 is commonly used to represent binary data as text so it can be safely included in JSON, HTML, CSS, URLs, email bodies, and API requests. You will often see it in data URLs, authentication headers, and file transfer payloads.
Is Base64 encryption?
No. Base64 is only an encoding format. Anyone can decode it easily, so it should not be treated as a security or privacy feature.
What is URL-safe Base64?
URL-safe Base64 replaces the standard + and / characters with - and _. This makes the encoded string safer to place in URLs, cookies, and filename-like contexts without extra escaping.
Why is my Base64 output longer than the original text?
Base64 expands the data because it uses 4 text characters to represent every 3 bytes of input. That means the encoded result is usually about one-third larger than the source data.