Data Storage Converter
Convert bits, bytes, KB, MB, GB, TB and binary KiB, MiB, GiB units instantly.
- Free, no account
- No watermark
- No usage limit
About the Data Storage Converter
Most storage converters ask you to pick a "from" unit and a "to" unit, then hand you back one number. That's fine, except "GB" doesn't name one fixed amount, and that mismatch is what sends people looking for a converter in the first place, so this one is built the other way round. Type a number, pick the unit you've got, and it prints that value in every unit at once, straight down one column. Bits and bytes sit at the top. Below them the decimal units stack together (KB, MB, GB, TB, PB), and below those the binary ones (KiB, MiB, GiB, TiB, PiB). GB lands in that decimal block and GiB a few rows further down in the binary one, both on screen together, so you read down the column and watch the same amount of storage turn into two different numbers. Everything runs locally in your browser, nothing is uploaded, and after the page loads it keeps converting with the Wi-Fi off.
That all-at-once view is the whole reason to use it. Storage genuinely isn't counted one way. Drive makers count in decimal, so a 1 GB drive is 1,000,000,000 bytes. Your operating system counts in binary, so its "GB" is really 1,073,741,824 bytes, and it prints "GB" on the label anyway, while networking people quietly switch to bits underneath you. A pick-one-then-pick-another converter buries all of that behind a single result, and you walk away trusting a number you never got to check. Put every unit on screen together and the mismatch is impossible to miss.
How to use
- Type your number in the Value box. Any size works, from a single bit up to thousands of petabytes.
- Pick the unit you have in the From dropdown. GB for a drive spec, MiB for something a Linux terminal reported, Bytes for an exact file size you copied.
- Read the whole table. Every unit updates live as you type, so you're not converting one guess at a time. Your source unit stays highlighted so you don't lose track of it.
- Copy any row with its copy button, or hit Copy shareable link to send someone the exact value and unit you're looking at, baked right into the link.
Blank the box or type something that isn't a number and every row just shows a dash, no NaN and no red error. A whole number of bits or bytes prints in full, every digit of it, so 1 PiB reads out as 1,125,899,906,842,624 bytes rather than getting rounded off. Values that land between whole units and run very small (below about a ten-thousandth) flip to scientific notation so they stay readable.
Why your "1 TB" drive shows 931 GB
The drive is fine, nothing is being hidden from you.
The manufacturer sold you 1 TB meaning exactly 1,000,000,000,000 bytes. That's an honest terabyte by the official SI definition, decimal, base 1000. When Windows looks at that same drive it divides those bytes by 1,073,741,824 to get its "GB" figure, because what Windows calls a GB is really a GiB (base 1024). Do that division, 1,000,000,000,000 divided by 1,073,741,824, and you land on about 931.
The gap scales with size, which is why it barely shows on a small flash drive and is obvious on a big disk. KB and KiB differ by about 2.4%. GB and GiB by roughly 7.4%. By the TB versus TiB level it's just under 10%. So a 500 GB SSD reads near 465, a 2 TB drive shows around 1.82 TiB, and the pattern holds all the way up. There is a second, much smaller reason a fresh drive shows slightly less than the box too, the file system reserves a little space for its own record-keeping, but that's a fraction of a percent. The 931-versus-1000 gap is almost entirely the decimal-versus-binary thing, and the converter shows you both figures so you can see precisely which one your source was using.
The plain names (KB, MB, GB, TB) are decimal and show up on boxes and in marketing. The "-bi-" names (KiB, MiB, GiB, TiB) are binary and show up in careful technical output. The IEC standardized those binary names back in 1998 specifically to end this argument, and mostly failed, because operating systems kept printing "GB" while meaning GiB.
Bits, bytes, and why a fast plan downloads slow
There's a second unit mix-up under all of this, and it's bits versus bytes. A bit is a single 1 or 0. A byte is eight of them, enough for one character. That factor of 8 is easy to miss because the only clue is usually capitalization: lowercase b means bits, uppercase B means bytes. Internet plans are quoted in bits (Mbps, megabits), files are measured in bytes (MB), and the two camps rarely announce which one they mean.
So a 100 Mbps plan doesn't download at 100 MB/s. Divide by 8 and the real ceiling is about 12.5 MB/s. The plan isn't lying, the unit just changed hands between the network and your computer. To sanity-check a download, get both sides into bytes first: convert the speed with Mbps divided by 8, drop the file size into the same byte unit, then divide one by the other. This tool converts the sizes for you, it doesn't do the rate math, but lining the units up correctly is most of the battle.
Frequently asked questions
Does the converter round the numbers?
Not when the answer is a whole number of bits or bytes. Storage is counted rather than measured, so a count prints every digit it has: 1 GiB reads 1,073,741,824 bytes, 1 PiB reads 1,125,899,906,842,624, and the Bytes row hands back exactly the figure you typed into the box. Anything that lands between whole units, a single byte read as gigabytes for example, shows eight significant figures, and values below about a ten-thousandth switch to scientific notation. Counts above roughly 9 quadrillion switch too, because that is the point where a browser's own number type stops being able to tell one whole number from the next and any digits past it would be invented. Every conversion routes through the byte using exact factors, so nothing lossy piles up along the way.
Is "16 GB of RAM" decimal or binary?
Binary, always, even though the box says GB. Memory is addressed in powers of 2 by how the chips are physically wired, so a stick is really 16 GiB (17,179,869,184 bytes), never a round decimal 16 billion. This is the one place the 1024 convention is correct because of physics rather than habit. Disks are the opposite, sold by the decimal byte count, which is a big part of why a RAM number and a disk number never seem to use the same "GB."
Why doesn't my download ever hit the full MB/s?
Because dividing by 8 only gives you a ceiling. Part of every connection carries overhead instead of your file, the packet headers and handshakes that make the transfer work, so even a flawless link runs a few percent short of the theoretical rate. Then the practical stuff stacks up, the server limiting how fast it sends, other devices sharing your line, Wi-Fi dropouts, even a drive that can't write incoming bytes as fast as they arrive. So a plan the math caps near 12.5 MB/s usually shows a little less in real use. The conversion tells you the most you could get, never a guarantee of what you will.
Can I share a specific conversion with someone?
Yes. The Copy shareable link button drops your current value and source unit into the URL, so whoever opens it lands on the exact same setup instead of you explaining "type 750 and pick GiB." Useful in a support thread or a group chat where somebody keeps misreading a drive size.
Why does the same drive show more free space on a Mac than on Windows?
Because the two operating systems count storage differently. macOS switched to decimal units back in Mac OS X 10.6, so it treats a GB as an even 1,000,000,000 bytes, the same way the drive maker does, and a 1 TB disk reads close to 1,000 GB. Windows still counts in binary and prints "GB" anyway, so that identical disk reads about 931 GB instead, without a byte on it having moved. Plug one external into both machines, or dual-boot, and the free-space figure jumps purely on which convention each side uses to describe it.
What's the smallest and largest it handles?
From a single bit at the low end up through thousands of petabytes at the high end, with everything in between. That range covers a tiny config file, a phone's storage, a full drive, and a data-center backup without switching tools. The scientific-notation fallback is what keeps the extreme ends of that range usable instead of unreadable.