Blog · Guides

40 QR Code Versions Every Designer Must Map to Print Sizes

For designers and printers: map 40 QR Code versions (21×21–177×177) to practical print sizes, ECC choices, and verification steps to avoid failed scans.

By AriaQR Team · · Published · 9 min read

40 QR Code Versions Every Designer Must Map to Print Sizes

Abstract QR print size mapping illustration

There are 40 standard QR code versions in the Model 2 family, and each one defines a specific grid size using the formula modules per side follow the formula (4 times the version number) plus 17. Version 1 starts at 21×21 modules and Version 40 reaches 177×177 modules. The version you pick sets a ceiling on data capacity, but how much you can actually fit depends on the data mode and the error correction level you choose.

TL;DR:

  • The maximum standard QR code size is Version 40, measuring 177×177 modules, which supports the highest data capacity but is harder to scan at small sizes.
  • Larger versions require higher print resolution and a sufficient quiet zone, as each module shrinks to fit the physical space for reliable scanning.
  • Capacity varies significantly depending on the data mode and error correction level, with maximums at Version 40 Level L, but actual payloads usually fall short of this theoretical limit.
  • Micro QR codes and variants like rMQR and iQR are suited for space-constrained or specialized industrial use, but for consumer scanning, Model 2's standard sizes and support are most reliable.
  • Testing printed codes on multiple scanners and at the smallest expected device size reduces the risk of scan failure in real-world conditions.

Table of Contents

How version numbers map to module dimensions

A "version" is not a code type or a payload format. It is simply the grid size of the symbol, nothing more. Version 1 is the smallest standard QR code at 21×21 modules, and every version above it grows by exactly 4 modules per side. That pattern comes straight from the formula: modules per side follow the formula (4 times the version number) plus 17.

Run the numbers and the logic becomes clear:

  • Version 1: (4 × 1) + 17 = 21 modules per side
  • Version 10: (4 × 10) + 17 = 57 modules per side
  • Version 40: (4 × 40) + 17 = 177 modules per side

Each step adds a ring of modules around the previous grid, which is why the standard caps out at 40 versions rather than scaling indefinitely. Beyond that point, symbols become too dense for most cameras and printers to resolve reliably.

Module counts and what they mean for printing and scanning

Knowing the module count for a given version tells you how fine the printed detail needs to be, and that has direct consequences for both print production and scan reliability.

  1. Version 1 (21×21) fits simple payloads like short URLs and scans reliably even at small print sizes.
  2. Version 2 (25×25) and Version 3 (29×29) add modest capacity for slightly longer text or contact data.
  3. Version 10 (57×57) suits payloads like vCards or structured data that need more room.
  4. Version 25 (117×117) is dense enough for large text blocks but demands careful print quality.
  5. Version 40 (177×177) holds the maximum data defined by the standard, but the fine module structure makes it unforgiving at small sizes.

Every QR code also needs a quiet zone, a blank margin around the symbol, for a scanner to locate it correctly. As module count rises, each module must shrink to keep the code a reasonable physical size, which raises the print DPI you need to keep edges crisp. A high-version code printed at low resolution tends to blur adjacent modules together, and that is one of the most common causes of failed scans in the field.

Data capacity by version: modes and error correction trade-offs

Capacity is not a single number tied to a version. It shifts with the encoding mode and the error correction level (ECC) you apply.

QR codes support four main data modes, each suited to a different kind of content:

  • Numeric mode handles digits only and packs the most characters per module.
  • Alphanumeric mode covers digits, uppercase letters, and a limited symbol set.
  • Byte mode stores raw binary data, including most URLs and mixed-case text, at a lower character density.
  • Kanji mode encodes double-byte characters efficiently for Japanese text.

Developer guides summarize capacity by mode and note that Version 40 at ECC level L gives the maximum theoretical capacity for each mode, though real-world capacity is usually lower once mixed content and formatting overhead are factored in.

Version 40 at ECC level L represents the ceiling of the standard QR symbology, and it is the figure most developer references cite when describing maximum theoretical capacity. Treat it as a reference point rather than a planning number. Real payloads rarely hit that ceiling cleanly, so testing your actual content against a target version is more useful than relying on published maximums.

Error correction levels trade capacity for resilience:

  • Level L restores a smaller share of damaged data and maximizes usable capacity, good for clean digital-only use.
  • Level M offers a balanced middle ground for general print use.
  • Level Q adds more redundancy for codes exposed to wear or handling.
  • Level H sacrifices the most capacity but tolerates the heaviest damage or logo overlays.

You can see how these levels interact with capacity in practice using our error correction demo, which shows the resilience difference at each level side by side.

QR code variants: Micro QR, rMQR, iQR, and Model differences

Standard Model 2 QR codes are not the only format in the family, and some projects call for an alternative shape or size entirely.

  • Micro QR codes range from 11×11 to 17×17 modules and suit space-constrained marking like small parts or labels, though their capacity is limited.
  • rMQR (rectangular Micro QR), specified under ISO/IEC 23941:2022, fits narrow canvases where a square code would not work, though it reduces compatibility with some native phone camera decoders.
  • iQR codes offer denser, sometimes rectangular layouts for specialized industrial use, but they lack the universal native support that standard QR codes enjoy.
  • Model 1, the original specification, was effectively superseded by Model 2, which added version information blocks and became the practical standard for open, public-facing systems.

For anything meant to scan on a stranger's phone, Model 2 remains the safer choice precisely because of that universal support.

Choosing a version, an ECC level, and a print size that actually works

Most everyday applications, from business cards to product packaging, never need more than Version 1 through 7. Those versions balance enough capacity for typical URLs or contact details with strong scan reliability across old and new devices alike. Save higher versions for genuinely large payloads, like embedded documents or archival data, where machine-to-machine reading matters more than consumer scanning.

A few practical rules keep codes reliable once they leave the screen and hit print:

  • Keep each module at least 2 to 3 pixels wide at your target scan distance to avoid blur.
  • Maintain a quiet zone of at least 4 modules around the symbol.
  • Use our size calculator to convert a module count into the right print dimensions and DPI for your job.
  • Choose ECC level Q or H when a logo sits on top of the code or when the print surface is likely to get scuffed or dirty.

Pro Tip: Always test your final code on more than one decoder app and on the smallest device your audience is likely to use, since screen size and camera quality both affect real-world scan success.

Why verification matters more than the spec sheet

Standards like ISO/IEC 18004:2024 tell you what a QR code is supposed to do under ideal conditions, but printed materials rarely stay ideal. Lighting changes, ink bleeds, and phone cameras vary wildly in focus distance and sensor quality, so a code that passes spec on paper can still fail in someone's hand.

Printed QR verification under real conditions

That gap is exactly why we built AriaQR around verification rather than generation alone. Every code we produce gets checked on real decoders before download, so you are not left guessing whether your chosen version and ECC level will actually hold up once it is printed. We also let you embed your own artwork into the code itself without breaking scan reliability, which matters more as version and module density increase.

If you are deciding between versions or ECC levels, run your draft through our size calculator and error correction demo before committing to a print run. Testing against your actual printer, paper stock, and target device closes the distance between a theoretical spec and a code that works on the first scan.

— Tharun

FAQ

How many versions of QR codes are there?

There are 40 standard QR code versions in the Model 2 family, ranging from Version 1 at 21×21 modules to Version 40 at 177×177 modules. Each version adds 4 modules per side following the formula modules per side follow the formula (4 times the version number) plus 17.

What are the different types of QR codes?

Beyond the standard Model 2 QR code, the family includes Micro QR (11×11 to 17×17 modules) for small-surface marking, rMQR for rectangular layouts, and iQR for denser industrial use. Model 1 was the original specification, but Model 2 became the practical standard because of its version information blocks and wider device support.

What is the latest version of a QR code?

Version 40 is the largest and highest-capacity version in the standard QR code symbology, holding the maximum module grid of 177×177. The underlying specification itself was most recently updated as ISO/IEC 18004:2024, which refined encoding efficiency and error correction behavior rather than adding new versions.

How do I know what type of QR code I have?

For Versions 7 through 40, the symbol includes built-in version information blocks that let a decoder identify the version automatically. For versions below 7, you generally need to count the modules per side and apply the formula, since no explicit version data is encoded in the symbol.

Sources

Recommended