Binary to Text

Paste the blob somebody sent you and find out what it says — or why it will not decode

Runs 100% in your browser•Nothing is uploaded to a server•Instant results

Direction

Base

One group per

Separator

UTF-8 byte is what a computer means and round-trips anything. Character is what a puzzle means. Pure ASCII is identical either way.

Input

When a binary string refuses to decode, it is almost always one of four things

Decoding is where binary gets interesting, because encoding cannot really fail and decoding fails constantly. The string arrives from a puzzle, a forum post, a game, a tattoo photo or a homework sheet, and the four ways it goes wrong are worth knowing by sight.

The length is not a multiple of eight. Count the digits. If the total does not divide by eight, a digit was lost in the copy or the sender was using seven-bit groups, which was normal in the era when ASCII was seven bits and the eighth was a parity check. A stray leading zero disappearing from the front of a group is the single most common transcription error, because it looks like nothing.

It is not actually binary. If there is a digit above 1 anywhere, the string is decimal or hex. Long runs of digits with the occasional letter a to f are hex — switch the base rather than trying to fix the string. This page names the offending group rather than returning an empty box, because “nothing happened” tells you nothing about which of eight hundred characters was wrong.

The groups are the wrong width. Unseparated binary is only decodable if you know the width. Eight is the usual answer and the one assumed here. If eight produces plausible-looking gibberish — readable letters in the wrong order, or every character shifted — try reading it as code points instead, which is how binary written for a puzzle rather than for a computer is usually built.

The text was never ASCII. If the decode produces the right number of characters and half of them are replacement marks, the original was UTF-8 with multi-byte characters and something truncated it, or the groups were code points and you are decoding as bytes. The mode toggle is the fix; those two readings are genuinely different for anything above 127.

Five groups, one word

01001000 01100101 01101100 01101100 01101111

Out

Hello

ASCII in three bases

If a decode looks close but wrong, comparing a few groups against this table usually shows the offset immediately.

Character	Binary	Hex	Decimal
(space)	00100000	20	32
!	00100001	21	33
"	00100010	22	34
#	00100011	23	35
$	00100100	24	36
%	00100101	25	37
&	00100110	26	38
'	00100111	27	39
(	00101000	28	40
)	00101001	29	41
*	00101010	2A	42
+	00101011	2B	43
,	00101100	2C	44
-	00101101	2D	45
.	00101110	2E	46
/	00101111	2F	47
0	00110000	30	48
1	00110001	31	49
2	00110010	32	50
3	00110011	33	51
4	00110100	34	52
5	00110101	35	53
6	00110110	36	54
7	00110111	37	55
8	00111000	38	56
9	00111001	39	57
:	00111010	3A	58
;	00111011	3B	59
<	00111100	3C	60
=	00111101	3D	61
>	00111110	3E	62
?	00111111	3F	63

Character	Binary	Hex	Decimal
@	01000000	40	64
A	01000001	41	65
B	01000010	42	66
C	01000011	43	67
D	01000100	44	68
E	01000101	45	69
F	01000110	46	70
G	01000111	47	71
H	01001000	48	72
I	01001001	49	73
J	01001010	4A	74
K	01001011	4B	75
L	01001100	4C	76
M	01001101	4D	77
N	01001110	4E	78
O	01001111	4F	79
P	01010000	50	80
Q	01010001	51	81
R	01010010	52	82
S	01010011	53	83
T	01010100	54	84
U	01010101	55	85
V	01010110	56	86
W	01010111	57	87
X	01011000	58	88
Y	01011001	59	89
Z	01011010	5A	90
[	01011011	5B	91
\	01011100	5C	92
]	01011101	5D	93
^	01011110	5E	94
_	01011111	5F	95

Character	Binary	Hex	Decimal
`	01100000	60	96
a	01100001	61	97
b	01100010	62	98
c	01100011	63	99
d	01100100	64	100
e	01100101	65	101
f	01100110	66	102
g	01100111	67	103
h	01101000	68	104
i	01101001	69	105
j	01101010	6A	106
k	01101011	6B	107
l	01101100	6C	108
m	01101101	6D	109
n	01101110	6E	110
o	01101111	6F	111
p	01110000	70	112
q	01110001	71	113
r	01110010	72	114
s	01110011	73	115
t	01110100	74	116
u	01110101	75	117
v	01110110	76	118
w	01110111	77	119
x	01111000	78	120
y	01111001	79	121
z	01111010	7A	122
{	01111011	7B	123
\|	01111100	7C	124
}	01111101	7D	125
~	01111110	7E	126

Binary to Text questions

It decoded but the text is gibberish. What now?

Check the group width first — the commonest cause is a string built as 7-bit ASCII being read as 8-bit, which shifts everything after the first character. Then try code points mode, which reads each group as one whole character rather than one UTF-8 byte. If the gibberish is readable letters in a scrambled order rather than symbols, the string is text but it was reversed or shifted, and it is a cipher rather than an encoding problem.

Does it handle binary with no spaces?

Yes. Unseparated input is chopped into eight-digit groups, which is the right guess for the overwhelming majority of what people paste. Commas, line breaks and mixed whitespace all work as separators too. What cannot be guessed is a string of seven-bit groups run together with no separator, because nothing in the digits themselves says where one character stops.

Why does one emoji come out as four odd characters?

Because you are decoding as code points and the source was UTF-8 bytes. An emoji is four bytes in UTF-8, and reading those four bytes as four separate characters gives you four meaningless ones instead of the emoji. Switch to UTF-8 bytes mode and the four groups recombine into the single character they encode.

Binary to Text

Direction

Base

One group per

Input

Output

When a binary string refuses to decode, it is almost always one of four things

ASCII in three bases

Binary to Text questions

It decoded but the text is gibberish. What now?

Does it handle binary with no spaces?

Why does one emoji come out as four odd characters?