Decode Base64 JSON
online, free & instant

From Base64 string to readable JSON in seconds

Private, instant, and zero dependencies.

Related JSON tools

Common questions answered

What is Base64 encoding for JSON?

Base64 encoding converts binary or text data into ASCII characters safe for text-based protocols. When applied to JSON, Base64 lets you embed a JSON document inside environments that do not support special characters — email headers, URL parameters, XML attributes, or other JSON strings.

This tool provides both encoding (JSON → Base64) and decoding (Base64 → JSON). Paste a JSON document to get its Base64 representation, or paste a Base64 string to decode it back. The decoder handles URL-safe Base64 (using - and _ instead of + and /) and padding variations. This is particularly useful for inspecting JWT payloads and encoded configuration strings.

{"user":"Alice","role":"admin"}

Encoded:
eyJ1c2VyIjoiQWxpY2UiLCJyb2xlIjoiYWRtaW4ifQ==

Decode the string above to get
the original JSON back.

Get the most out of this tool

• JWT tokens contain Base64URL-encoded JSON in their header and payload — decode them here to inspect the claims.
• Use Base64 encoding when embedding JSON in URL query parameters to avoid escaping issues with special characters.
• Base64 increases size by approximately 33% — use it only when transport requires ASCII-safe encoding, not for compression.

Understanding Base64 encoding of JSON data

Base64 is a binary-to-text encoding scheme that represents arbitrary binary data using a 64-character alphabet: A–Z, a–z, 0–9, plus (+), and slash (/), with equals signs (=) as padding. JSON is a text format, but when JSON must be embedded in contexts designed for binary data — like binary protocols, URL parameters, or HTTP headers — Base64 encoding converts the JSON text to a safe, ASCII-only representation that survives transmission through systems that are not 8-bit clean.

The JWT (JSON Web Token) standard is the most common context where developers encounter Base64-encoded JSON. A JWT consists of three Base64URL-encoded parts separated by dots: the header (algorithm and token type), the payload (claims like user ID and expiration), and the signature. Base64URL is a variant of Base64 that replaces + with - and / with _, producing URL-safe strings without percent-encoding.

Base64 encoding increases data size by approximately 33%. Every 3 bytes of input produce 4 Base64 characters. A 300-byte JSON object becomes roughly 400 characters when Base64-encoded. This size overhead is the primary downside of Base64 — for large JSON payloads, the encoding overhead is significant. Size-sensitive applications should prefer alternative encoding schemes or avoid embedding large JSON in Base64 contexts.

URL-safe Base64 (Base64URL) is critical when Base64-encoded JSON appears in URLs, URL parameters, or HTTP headers. Standard Base64 uses + and / characters that have special meanings in URLs — + means space in URL-encoded query strings, and / is a path separator. Base64URL substitutes - and _ respectively, producing strings that can be included in URLs without percent-encoding and without altering the decoded value.

Kubernetes stores secrets as Base64-encoded values in YAML manifests. A Kubernetes Secret resource contains a data map where values are Base64-encoded strings — for example, a database password is stored as cGFzc3dvcmQ= rather than password. When Kubernetes secrets contain JSON-formatted values (like service account credentials or Docker registry configurations), decoding the Base64 reveals the nested JSON that needs inspection and formatting.

Environment variables in containerized environments sometimes contain Base64-encoded JSON configuration. Container orchestration platforms like Kubernetes and cloud functions platforms (AWS Lambda, Google Cloud Functions) support injecting configuration as environment variables. When a complex configuration object must be passed as a single environment variable, Base64-encoding the JSON produces a single string that can be stored in one environment variable and decoded at runtime.

The browser's atob() and btoa() functions handle Base64 decoding and encoding. atob("SGVsbG8=") returns "Hello". However, these functions handle only Latin-1 characters — Base64-encoded UTF-8 strings containing non-ASCII characters (emoji, accented characters, CJK characters) require additional UTF-8 decoding after Base64 decoding. Modern browser environments provide TextDecoder for proper UTF-8 handling: new TextDecoder().decode(Uint8Array.from(atob(b64), c => c.charCodeAt(0))).

When Base64 JSON decoding is needed

Base64 JSON decoding pitfalls

• Confusing standard Base64 with Base64URL: Standard Base64 uses + and /; Base64URL uses - and _. If you try to decode a Base64URL string (like a JWT segment) with a standard Base64 decoder, the substituted characters cause decoding errors. Always check which variant your source uses before decoding.
• Forgetting UTF-8 decoding for non-ASCII content: Base64 decodes bytes, not characters. If the original JSON contained non-ASCII characters (emoji, accented letters, CJK), the decoded bytes need UTF-8 interpretation to produce the correct string. Using atob() alone without UTF-8 decoding produces garbled output for non-Latin-1 content.
• Missing padding characters: Standard Base64 requires padding with = characters to make the length a multiple of 4. Base64URL strings often omit padding. If a Base64 decoder returns an error about invalid padding, try adding one or two = characters to the end of the string before decoding.
• Treating Base64 as encryption: Base64 is encoding, not encryption. Anyone who sees a Base64-encoded string can immediately decode it without any key. Never use Base64 to "hide" sensitive values — use proper encryption algorithms for that purpose. Kubernetes secrets stored as Base64 are not encrypted; they require Kubernetes encryption-at-rest to be truly secret.

Additional frequently asked questions