Tool AT-W9REV. 2026-A

Password generator for secure passwords and passphrases

A secure password is above all long and random, and it is used for one account only. This password generator creates it in your browser: as a passphrase of four to eight words, as a random password of 8 to 64 characters, as a PIN or as a Wi-Fi key. Nothing is transmitted or stored. Each suggestion comes with its strength, calculated exactly in bits.

  • free
  • no sign-up
  • stays in your browser

Generate a password or passphrase

Generator · five suggestions per run

The suggestions are created in your browser and are neither transmitted nor stored. With your next click, or when you close the page, they are gone. So copy the suggestion you choose straight into your password manager.

Step 1What do you need?
Step 2Set up the passphrase

If a service demands capital letters, digits and special characters: capitalising each word, appending a number and using a hyphen as the separator meet all three. The word list contains only letters from a to z, without umlauts, so that the passphrase works on any keyboard.

crypto.getRandomValues · nothing leaves this browser

Suggestions

  1. The suggestions appear here as soon as the page has finished loading.

Strength

—

nothing generated yet

Offline guessing, on average

—

Assumption: 10 billion guesses per second after a data breach. A simplified model for comparison, not a prediction.

How is a password created that nobody can predict?

The randomness comes from your browser. Every current browser has a source of randomness built in for encryption, the interface is called crypto.getRandomValues, and that is exactly what the generator uses. The ordinary random function of JavaScript is not suitable for this. It is built for games and animations, not to make sure nobody can predict its next number.

The source returns numbers between zero and just over 4.29 billion. To turn one of these into one of 94 characters, you could simply take the remainder of a division. That is almost fair, but only almost: 4.29 billion does not divide evenly by 94, so a few characters would turn up by a tiny fraction more often. The generator therefore discards the few numbers at the top end that form the overhang, and draws again. With the 2,048 words the sum works out evenly anyway.

If a random password is to contain every chosen character type, the generator does not simply insert one character of each type. That would skew the distribution. It generates the whole password again until every type is present, and the strength display deducts exactly this condition. That is why at 16 characters it shows slightly less than 16 times 6.6 bits.

The suggestions are created only on your device. The page sends them nowhere, does not keep them in browser storage and forgets them with your next click. The browser fetches the word list once from our server, like an image; which words are drawn from it, the server does not learn.

ALL IN THE BROWSERTHE BROWSER'SRANDOM SOURCEUNBIASEDSELECTIONFIVE SUGGESTIONSYOUR PASSWORDMANAGERALL IN THE BROWSERTHE BROWSER'SRANDOM SOURCEUNBIASED SELECTIONFIVE SUGGESTIONSYOUR PASSWORD MANAGER
FIG. 1 · HOW A SUGGESTION IS CREATEDGOLD = SAVE TO YOUR PASSWORD MANAGER

Why there is no field here for checking a password

Many websites offer to test the strength of an existing password. We do not, and that is deliberate. Your real password belongs in exactly two places: the service's sign-in field and your password manager. Any other site, ours included, would have to ask you to believe that it records nothing. From the outside that is hard to verify.

On top of that, with a password you made up yourself a check can only estimate. It counts character types but does not know the patterns attackers try first. How badly gut feeling misleads here is well documented Ur 2016Bonneau 2012. With a generated password, by contrast, the strength is known exactly, because chance sets it. Whether a stored password has turned up in a data breach is reported by the major browsers and most password managers on their own, without sending the password in plain text.

For the same reason our access for AI assistants (further down) does not generate passwords. A secret that is created through an AI service passes through other parties' systems and their logs. The assistant only gets the calculation of how strong a method is, and the pointer to this page.

How long does a secure password need to be?

What counts is the number of possible combinations, given in bits. Every extra bit doubles the work of guessing, ten more bits mean a good thousand times as much. One word from our list adds exactly 11 bits, one character from all 94 printable characters about 6.6 bits, one digit 3.3 bits.

From this follows the most important rule: length counts for more than special characters. A password of 20 characters made only of lower-case letters and digits, at 103.4 bits, is clearly stronger than one of 12 characters from all four character types (78.1 bits). Large studies using real guessing methods reach the same conclusion Kelley 2012Shay 2016.

Randomly generated: strength and rough guessing time
VariantCharactersBitOffline guessing, on averageAssessment
PIN, 6 digits619,9under a secondonly with lockout
Passphrase, 4 words≈ 2844about 15 minutesweak
Random password, 8 characters, all four character types851,3about 1.6 daysmedium
Passphrase, 5 words≈ 3555about 21 daysmedium
Passphrase, 6 words (default)≈ 4266about 120 yearsstrong
Random password, 12 characters, all four character types1278,1about 530,000 yearsstrong
Passphrase, 8 words≈ 5688about 490 million yearsvery strong
Wi-Fi key, 20 characters20100more than a trillion yearsvery strong
Random password, 20 characters, lower-case letters and digits only20103,4more than a trillion yearsvery strong
Random password, 16 characters, all four character types (default)16104,6more than a trillion yearsvery strong

The time column is a simplified model: someone has captured a service's stored password hashes and tries 10 billion passwords per second offline. For fast hashing methods that were never meant for passwords, that is a realistic order of magnitude. Services that store passwords with slow methods such as bcrypt, scrypt or Argon2 cut the rate by many powers of ten. With an online account that slows down after a few failed attempts, nobody gets anywhere near. So the table shows the unfavourable case. That is the one we mean, because you cannot see from the outside whether a service stores your data properly.

For an ordinary account, a passphrase of four words is therefore enough, as long as the service slows down failed attempts. For the master password of your password manager, use six words or more: whoever captures its file can attack it offline. Passwords people make up themselves, incidentally, fall far below every row of this table. An analysis of almost 70 million accounts came to about 20 bits offline Bonneau 2012.

What the BSI and NIST recommend

  • BSI: at least 8 characters from four character types, or at least 25 characters from two. For Wi-Fi, at least 20 characters. As a passphrase, the BSI names five to six randomly chosen words. Each password is for one account only.
  • BSI on changing passwords: change one if there are signs that an unauthorised person knows it, for example after a data breach or after malware has been on the device (clean the device first). The BSI dropped routine changes from the IT-Grundschutz Compendium with the 2020 edition. Research had shown that replacement passwords can often be derived from the old ones Zhang 2010.
  • NIST SP 800-63B-4 (US guideline, July 2025): passwords that protect on their own, at least 15 characters; with a second factor, at least 8. Services should allow at least 64 characters, must not impose composition rules such as “at least one special character”, and must not demand regular changes. Password managers must work, and pasting should be allowed.

Information checked on 13 November 2026. Links to the BSI and NIST are below under the sources.

Passphrase or random password: which suits what?

The same number of bits is equally strong, whether it comes from words or from characters. The difference lies in everyday use, and it is smaller than is often claimed. In a study with almost 1,500 participants, assigned passphrases were no easier to remember than equally strong random passwords, and typing them took longer Shay 2012.

The passphrase still has its place: wherever you type, read out or dictate it yourself. That includes the master password of the password manager, signing in to your Windows or Mac account, and the password for your e-mail inbox, through which most other accounts can be reset. On a phone you type words without switching to the symbol keyboard, and on the telephone nobody asks whether that was a capital or a lower-case O.

All that matters is that chance chooses the words and not you. Word sequences you put together yourself follow the language, often they are well-known titles or sayings, and those rank near the top of every attack list Bonneau & Shutova 2012.

For everything the password manager fills in for you, use a random password. You never have to remember it, so it can be short and dense: 16 characters from all character types come to 104.6 bits and also fit into forms with a length limit.

506480Passphrase, 4 words44 bits · ≈ 28 charactersRandom password, 8 characters51.3 bits · ≈ 8 charactersPassphrase, 6 words66 bits · ≈ 42 charactersRandom password, 12 characters78.1 bits · ≈ 12 charactersPassphrase, 8 words88 bits · ≈ 56 charactersRandom password, 16 characters104.6 bits · ≈ 16 characters
Fig. 2 · Strength in bits, with typical lengthOutline = passphrase · filled = random password · lines: 50 / 64 / 80 bits

Where to put all those passwords?

A separate password for every account: most people fail at this rule, understandably. Anyone with many accounts gets by through reuse Pearman 2017, and between 42 and 51 per cent of users use the same password on several sites Das 2014. Once it has leaked, attackers try it everywhere. A slight variation helps little: whoever knows the old one often guesses the new one in a few attempts Wang 2016. A password manager solves this. You remember one, and it remembers the rest.

Built in

Browser or device

Chrome, Edge, Firefox and Safari store passwords and fill them in; on iPhone, iPad and Mac there is the Passwords app for this. It costs nothing and is often already set up. The BSI points out that malware can read credentials out of browsers comparatively easily; an up-to-date system and a device lock are therefore part of the deal.

good for getting started

Local

KeePassXC

A free programme for Windows, macOS and Linux. It keeps all passwords in an encrypted file on your computer, and no account with a provider is needed. Anyone using the file on several devices has to take care of syncing and backing up themselves.

good if your data is to stay in-house

Synced

Bitwarden

A service with open-source code that brings the encrypted vault to all your devices via the provider's servers, with apps for the phone and extensions for the browser. The basic functions cost nothing; for families and teams there are paid plans with shared vaults.

good for several devices and people

The three stand for three designs, not for a ranking; 1Password, Proton Pass and others work on the same principle. More important than the product is how you use it. Whether a manager improves things depends on how it is used from the moment a password is created Lyastani 2018. So let it generate new passwords instead of just copying the old ones into it.

The master password and the emergency sheet

The password manager needs a password that you really remember: a passphrase of six words or more, which you use nowhere else. Writing it down is allowed if the note is kept under lock and key, says the BSI; it does not belong on the screen. Alongside it belong the recovery codes for the password manager and the e-mail account. If the master password is lost, in the worst case all stored logins are gone.

So that your relatives know where this note is in an emergency, the digital legacy planning sheet helps. What else belongs with it is in the guide on the digital legacy. And if you are standing in front of an account you can no longer get into: the guide Forgotten password, lost access goes through recovery step by step.

Moving over without locking yourself out

  1. Set up the manager. With a new passphrase as the master password, written down and locked away. Print the recovery codes at the same time.
  2. E-mail account first. New random password, switch on two-factor sign-in. Almost all other accounts can be reset through this inbox, which is why it comes first.
  3. Then the important accounts: bank, mobile contract, shops with stored payment details, social networks. Each gets its own password generated in the manager.
  4. The rest next time you sign in. The manager offers to save the old password; swap it for a generated one while you are at it. After a few weeks almost everything has moved over this way.
  5. Destroy old lists. Notebook and text file on the desktop: get rid of them as soon as everything is in the manager. Delete the file from the recycle bin as well.

What helps if a password is stolen anyway?

Against a fake sign-in page even the strongest password does nothing: whoever types it in there hands it over. That is why the BSI advises a second factor for important accounts, above all the e-mail inbox. After the password, the service then also asks for a confirmation on your phone, a code from an app or a security key. Google has evaluated how much difference this makes across more than 350,000 real takeover attempts: a confirmation via a device belonging to the account holder stopped more than 94 per cent of attempts following phishing and all automated ones, whereas security questions sometimes stopped only about one in ten Doerfler 2019. The BSI rates hardware-based methods as particularly secure.

Passkeys: signing in without a password

A passkey replaces the password entirely. Your device creates a key pair for it: the secret part stays on the device or in the password manager, and the service receives only the public part. When signing in you confirm with a fingerprint, your face or the device PIN. Because the passkey is tied to the address of the service, it simply does not work on a copycat page, and nothing can be given away there. The BSI calls passkeys a gain in security and convenience and advises setting up a fallback, for example a second security key.

If a service offers you a passkey, there is little against it. In a large laboratory study, participants were willing to swap the password for a security key, but also had new concerns Lyastani 2020. The obvious question, what happens if the device is lost, is answered by the fallback. With many services the old password also remains as an emergency way in. It should still be strong.

Sharing passwords in a club and in the office

In a club and in a small office, several people often share one login: the club website or the customer account with the wholesaler. Such logins belong in a shared vault of a password manager, not in a chat group and not in a spreadsheet on the club laptop. The vault shows who has access and takes it away again with one click.

When someone gives up their role or leaves the company, change the passwords that person knew. This is one of the occasions when a change really is a must. How to do this properly is covered in the guide Employee joins or leaves; what else comes up in IT for a club, in the guide IT for clubs. For the guest Wi-Fi in the clubhouse, generate a Wi-Fi key above and turn it into a notice with the QR code generator.

What the research says

Six findings on which the defaults and advice of this page are based. The full details of each study are below under the sources.

Made up yourself

≈ 20 bits

That was the protection offered by the self-chosen passwords of almost 70 million accounts against offline guessing, and less than 10 bits online. No user group did noticeably better. Bonneau 2012

In the generator: chance chooses, which is why the bit figure is right.

Length before complexity

16 characters

“At least 16 characters” withstood large guessing attacks better than “8 characters with four character types and a dictionary check”. Length rules were often easier to handle, too. Kelley 2012Shay 2016

Default: 16 characters.

Reuse

42–51 %

of users used the same password on several sites. With one leaked password and a few personal details, researchers guessed the password of ordinary users within 100 attempts in over 73 per cent of cases. Das 2014Wang 2016

Hence: every password only once.

Passphrases

equally memorable

Assigned passphrases were no easier to remember than equally strong random passwords, and slower to type. Self-chosen word sequences are too predictable. Shay 2012Bonneau & Shutova 2012

In the generator: words by chance only, a passphrase for what you type.

Forced changes

derived

Replacement passwords after a forced change could be derived from the old ones with a purpose-built search method. The authors therefore call forced changes into question. Zhang 2010

Advice: change only when there is a reason.

Second factor

> 94 %

of takeover attempts following phishing failed against a confirmation via a device belonging to the account holder, and the automated ones all did. Doerfler 2019

Advice: a second factor for e-mail and bank.

What the generator does not do

  • It does not protect a device that already has malware on it. Whoever is reading along sees even the best password. What to do then is on the page Viruses and security.
  • It does not know your service's rules. If a form rejects a suggestion, adjust the length or character types and generate again.
  • It remembers nothing. What you have not copied is gone after the next click. What you copied sits in the clipboard; if a history is switched on there (on Windows, with the Windows key and V), delete the entry after pasting.
  • The time figures are a model. They show how the variants compare with one another, not when a particular account will fall.

Using this tool from an AI assistant

Assistants such as Claude or ChatGPT that speak the Model Context Protocol (MCP) can call this page’s calculation directly instead of rebuilding it. The server is public, needs no sign-in and stores nothing. The answer always includes the link to this page, the date of the data and the limits of the result.

Endpoint
https://aureum-tech.com/mcp
Tool
werkzeug_passwortstaerke
Transport
Streamable HTTP, read-only
Inputs
NameRequiredMeaning
verfahrenyes“zeichen” for a random password of individual characters, “woerter” for a passphrase of randomly drawn words.
zeichenvorratnoOnly for “zeichen”: number of possible characters per position, e.g. 10 (digits), 26 (lower-case letters), 62 (letters and digits), 94 (all printable ASCII characters).
laengenoOnly for “zeichen”: number of characters.
wortlistenoOnly for “woerter”: size of the word list drawn from, e.g. 2048 (this generator's list) or 7776 (the usual dice list).
woerternoOnly for “woerter”: number of words.
zwecknoWhat the password is meant for: online account, master password of a password manager, Wi-Fi key or PIN. Controls the recommendations.

Questions where an assistant should call it

  • “How many words does a secure passphrase need?”
  • “Is a 12-character password of letters, digits and special characters still secure enough today?”
  • “How long should my Wi-Fi password be?”
  • “How strong is a passphrase of five words from a dice list?”

How the connection works and what MCP can do in your own business is explained on the page about MCP integrations.

Common questions about passwords

How long should a secure password be?

For accounts protected by the password alone, the US guideline NIST SP 800-63B requires at least 15 characters. The BSI (the German Federal Office for Information Security) names two routes: at least 8 characters from four character types, or at least 25 characters from two. A generated random password of 16 characters comes to about 104 bits, a passphrase of six words from our list to 66 bits. What matters is that chance does the choosing.

Are the passwords from this generator secure?

They are created on your device using the browser's cryptographic random source (crypto.getRandomValues). Surplus random numbers are discarded so that every character and every word is exactly equally likely. The page transmits and stores nothing. A password only stays secure, though, if it is used for exactly one account and kept in a password manager rather than on a note stuck to the screen.

What is a passphrase?

A sequence of randomly drawn words, for example six words with a hyphen between them. Each word from our list of 2,048 words adds exactly 11 bits, so six words give 66 bits. What matters is that chance picks the words: a sentence you make up yourself follows the language and is therefore much easier to guess. Passphrases suit anything you type in yourself.

Do I have to change my passwords regularly?

No, not without a reason. The BSI dropped routine changes from its IT baseline protection (IT-Grundschutz) in 2020, and the US guideline NIST SP 800-63B forbids services from demanding them. You should change a password after a reported data breach, after malware has been on the device (clean it first, then change), or if someone knows it who should no longer know it.

Why can't I check my own password here?

Because you should not type your real password into any other website, ours included. From the outside it is hard to tell whether a page records what you type. Besides, a check on a password you made up can only estimate, because it does not know the patterns attackers try first. For a generated password the strength is known exactly.

What is a passkey?

A way of signing in without a password. Your device creates a key pair: the secret key stays on the device or in the password manager, and the service stores only the public one. You confirm with a fingerprint, your face or the device PIN. Because the passkey is tied to the address of the service, it does nothing on a fake page. The BSI advises setting up a fallback, for example a second security key.

How long should the Wi-Fi password be?

WPA2 and WPA3 allow 8 to 63 characters, and the BSI recommends at least 20. For this the generator uses lower-case letters and digits without easily confused characters such as l and 1. That is exactly 5 bits per character, so 100 bits for 20 characters. For guests, turn the key into a notice to scan with our QR code generator.

What the tool is based on

The guide values and notes above rest on these 13 scientific papers. Each has been checked via its DOI or the publisher’s page (last on 13 November 2026): it exists, it has not been retracted, and its finding still holds for the question today.

  1. Bonneau, J. (2012): The Science of Guessing: Analyzing an Anonymized Corpus of 70 Million Passwords. IEEE Symposium on Security and Privacy 2012, pp. 538–552

    Finding Self-chosen passwords from almost 70 million Yahoo accounts offer only about as much protection against offline guessing as 20 random bits, and less than 10 bits against online guessing. No user group did noticeably better, and graphical strength meters changed little.

    In the tool This is why the generator lets chance choose and not the person; only then is the bit figure shown correct.

  2. Kelley, P. G.; Komanduri, S.; Mazurek, M. L. et al. (2012): Guess Again (and Again and Again): Measuring Password Strength by Simulating Password-Cracking Algorithms. IEEE Symposium on Security and Privacy 2012, pp. 523–537

    Finding Among 12,000 passwords under seven rules, “at least 16 characters” withstood large guessing attacks best, better than “8 characters with four character types and a dictionary check”. Estimated entropy is only a rough rule of thumb for self-chosen passwords.

    In the tool Default of 16 characters for random passwords; the bit calculation expressly applies only to generated passwords.

  3. Shay, R.; Komanduri, S.; Durity, A. L. et al. (2016): Designing Password Policies for Strength and Usability. ACM Transactions on Information and System Security 18 (4), pp. 1–34

    Finding Two online studies with over 20,000 participants on 15 password rules: rules that rely on length rather than on character types can be both more secure and easier to handle.

    In the tool The text recommends length before complexity; in the generator, character types are only a means of meeting form rules.

  4. Shay, R.; Kelley, P. G.; Komanduri, S. et al. (2012): Correct horse battery staple: exploring the usability of system-assigned passphrases. Proceedings of the Eighth Symposium on Usable Privacy and Security (SOUPS 2012), ACM, pp. 1–20

    Finding In an online study with 1,476 participants, assigned passphrases of three or four words were no easier to remember than equally strong random passwords; entering them took longer and needed corrections more often.

    In the tool The page therefore does not promise effortless remembering, but recommends passphrases only where you type them yourself.

  5. Bonneau, J.; Shutova, E. (2012): Linguistic Properties of Multi-word Passphrases. Financial Cryptography and Data Security (FC 2012 Workshops), LNCS, Springer, pp. 1–12

    Finding Self-chosen word sequences in an Amazon passphrase system were anything but random: they favoured common pairs of two nouns, often film or book titles. Even three- or four-part phrases from natural language did not hold up against offline guessing.

    In the tool The generator draws the words of the passphrase at random from a fixed list, never from a sentence.

  6. Ur, B.; Bees, J.; Segreti, S. M. et al. (2016): Do Users' Perceptions of Password Security Match Reality?. CHI 2016, ACM, pp. 3748–3760

    Finding The 165 participants were clearly mistaken about how much security familiar sayings, appended digits and keyboard patterns add.

    In the tool One more reason not to offer a check field for your own passwords: gut feeling misleads, and so does an estimate.

  7. Das, A.; Bonneau, J.; Caesar, M. et al. (2014): The Tangled Web of Password Reuse. Network and Distributed System Security Symposium (NDSS) 2014, Internet Society

    Finding 42 to 51 per cent of the users examined used the same password on several websites. Anyone who varies it follows a few patterns that make guessing much easier for attackers.

    In the tool The text puts “each password for one account only” ahead of every length rule.

  8. Wang, D.; Zhang, Z.; Wang, P. et al. (2016): Targeted Online Password Guessing: An Underestimated Threat. ACM CCS 2016, pp. 1242–1254

    Finding If an attacker knows another, leaked password of the same person and a few personal details, he guesses the password of ordinary users within 100 attempts in over 73 per cent of cases.

    In the tool Explains why the online lockout alone does not protect if a password is used several times or in varied form.

  9. Pearman, S.; Thomas, J.; Naeini, P. E. et al. (2017): Let's Go in for a Closer Look: Observing Passwords in Their Natural Habitat. ACM CCS 2017, pp. 295–310

    Finding Measurement on the computers of 154 people over an average of 147 days: reuse is more common than assumed, especially partial reuse. People who have to manage many passwords get by with it; password managers possibly showed no effect on reuse or strength.

    In the tool The moving guide says: let the manager generate new passwords, do not copy old ones into it.

  10. Ghorbani Lyastani, S.; Schilling, M.; Fahl, S. et al. (2018): Better managed than memorized? Studying the Impact of Managers on Password Strength and Reuse. 27th USENIX Security Symposium 2018, pp. 203–220

    Finding Survey of 476 and browser measurement of 170 people: password managers change the security of passwords, for better or worse, depending on how they are used from the moment a password is created.

    In the tool Explains why the page links the generator with the manager instead of recommending the manager alone.

  11. Zhang, Y.; Monrose, F.; Reiter, M. K. (2010): The security of modern password expiration: an algorithmic framework and empirical analysis. ACM CCS 2010, pp. 176–186

    Finding Across more than 7,700 accounts: whoever has to replace an expired password often chooses something that can be derived from the old one. The authors therefore call forced changes into question in principle.

    In the tool Together with the BSI and NIST, supports the statement that passwords should only be changed when there is a reason.

  12. Doerfler, P.; Thomas, K.; Marincenko, M. et al. (2019): Evaluating Login Challenges as a Defense Against Account Takeover. The World Wide Web Conference (WWW 2019), ACM, pp. 372–382

    Finding Over 350,000 real takeover attempts at Google: prompts via a device belonging to the account holder stopped over 94 per cent of attempts following phishing and all automated ones; knowledge questions sometimes stopped only 10 per cent in the case of phishing.

    In the tool Backs the section on the second factor: a device as a second factor protects where the password has already been given away.

  13. Ghorbani Lyastani, S.; Schilling, M.; Neumayr, M. et al. (2020): Is FIDO2 the Kingslayer of User Authentication? A Comparative Usability Study of FIDO2 Passwordless Authentication. IEEE Symposium on Security and Privacy 2020, pp. 268–285

    Finding In a large laboratory study, participants were willing to replace the password with a security key (FIDO2), but named new concerns that may slow its adoption.

    In the tool Supports the advice to accept passkeys and to set up a fallback at the same time.

Standards, authorities and manufacturer data

Mon–Fri 8:00–17:00

We set up the password manager for you, for your family or office too, and move your existing logins across with you. This can usually be done remotely.

+49 3744 4399760[email protected]

Mon–Fri 8:00–17:00 · Straße der Jugend 5, 08228 Rodewisch

Have a password manager set up
+49 3744 4399760Mon–Fri 8:00–17:00
CallCallback

PARTS LIST · SHEET 1/1

What this page is made of

Values measured at build time. With JavaScript, your browser measures them again live.

ITEMPARTTYPESIZE
01index.htmlDocument≈ 29 kB
02styles.cssStylesheet≈ 20 kB
03scripts.jsScript≈ 13 kB
04archivo-variable.woff2Font≈ 58 kB
05ibm-plex-mono.woff2Font≈ 30 kB
06favicon.svgImage< 1 kB

Advertising or analytics cookies: none. Without your consent, the page loads nothing from third-party servers apart from Cloudflare, our delivery service. Fonts, scripts and images are hosted by us. That is how we build your IT, too.

OUTSIDE OFFICE HOURS

Feel free to call anyway

Our regular office hours are Monday to Friday, 8:00–17:00.

In the evening and at weekends the number is forwarded to us, and often someone will pick up. We cannot promise that, though.

If nobody answers, leave a short message on the voicemail: your name, your number and what it is about. We will call back as soon as we can, on the next working day at the latest.

The call costs nothing. If someone has to start work straight away outside office hours, the surcharge in the price list applies.

Call now+49 3744 4399760

Prefer to write? Use the contact form or e-mail [email protected]. We will read it on the next working day.

DIAGNOSIS · AT-K2

What do you need help with?

↵ SELECTESC CLOSECALL: +49 3744 4399760

Custom settings

Which services may load?

Strictly necessary

The choice you make here and your settings for colour scheme and motion. These are kept as small entries on your device and are never sent to a server. They are needed for the site to work and therefore cannot be switched off.