Omissis
ItalianoDownload
Beta · open source · Apache 2.0

Use artificial intelligence without handing it your data

Omissis replaces names, e-mail addresses, phone numbers, tax codes, IBANs and passwords with placeholders before the text leaves your computer, and puts the real values back into the answer. You read the real text. The service only sees placeholders.

Write to Giulia Ferraresi[PERSONA_1] (giulia.ferraresi@nevatek.example[EMAIL_1], mobile 347 5551234[TELEFONO_1]) at Nevatek S.r.l.[ORG_1]: the refund goes to IBAN IT60X0542811101000000123456[IBAN_1].

Made-up data. The replacement happens on your computer.

41kinds of data recognised: 28 always, 13 if you ask
0data sent to Omissis servers: there are none
26 msfor a short text on the CPU alone, no graphics card
2model sizes: 537 MB and 1.2 GB

How it works

Three steps, all on your computer. The AI service does not know Omissis exists.

You Omissis127.0.0.1 AI service real text placeholders only the answer real text
  1. 1

    Replace

    A model trained for Italian, plus checksum rules, finds personal data and replaces it with labels such as [PERSONA_1].

  2. 2

    Remember

    The link between label and value stays in a vault on your PC. The same person always gets the same label, even days later.

  3. 3

    Restore

    When the answer arrives, labels turn back into the real values: in the text, and also in the files and commands an agent writes for you.

If protection does not answer, the message does not leave. Never in clear by mistake.

Where it works

One engine, several ways to use it. Each has been verified to a different degree, and we say so.

lab-tested

In the browser

An extension for Chrome and Edge. It replaces the data right in the message box, before the site reads it, and re-checks what leaves: no proxy, no certificate. It also logs the AI services you visit and can block them.

verified live

In the terminal

A local address for Claude Code, Codex, opencode, aider and similar tools, subscription sign-ins included. It masks prompts, files read and past commands; it restores the text and whatever the agent writes. If you already have a gateway you keep it: Omissis sits in front of it.

lab-tested

From the keyboard

Ctrl+Alt+M masks the copied text, Ctrl+Alt+R restores the real values. Works with any program and any subscription.

lab-tested

With documents

Word, Excel, PowerPoint, PDF and whole folders: you give the AI the anonymised copy, formatting intact. Scans are read on your PC.

experimental

In desktop apps

Two ways, with the app used as usual and on your own subscription. The connector replaces what you type in the Claude, ChatGPT and Copilot apps for Windows: verified on the ChatGPT app, not yet on the others. The MCP server has Omissis read your documents, and the assistant only receives placeholders: for Claude Desktop, Cursor, Codex and the other apps that support it, tested live with Claude Code.

lab-tested

On the network

A proxy for programs that accept a local certificate, with an inventory of the AI tools on the PC and of the services contacted (Shadow AI).

Simple for those who use it, complete for those who manage it

The program opens in simple mode: status, text, documents and guide. Advanced mode adds the activity log, per-service policies and the inventory. The interface is in English and Italian.

The home page, as the person using the computer sees it
The home page, as the person using the computer sees it
Paste: the masked text is already on the clipboard, ready for the AI
Paste: the masked text is already on the clipboard, ready for the AI
A dropped document: the anonymised copy downloads by itself
A dropped document: the anonymised copy downloads by itself
Connecting terminal tools
Connecting terminal tools
One choice per service: protect, log, block
One choice per service: protect, log, block
Language, theme and signed updates
Language, theme and signed updates

The model

A classifier trained from mmBERT on about one million texts: 200,000 synthetic Italian documents written by our own generator and the training splits of nine public datasets, in 23 languages. Everything is synthetic except two collections of court decisions, which contain names that are already public in the decisions.

omissis-baseomissis-small
Parameters307.8 M140.9 M
Size1.2 GB537 MB
Time for a short text, on CPUabout 61 msabout 26 ms
Known sensitive values that end up masked*98.8 %98.8 %
Share of text masked*17.0 %17.2 %
Company names in free text masked*94 %92 %
Public benchmark PIIMB, average score**0.9110.912

* On a private set of 2,000 real IT help-desk texts, not releasable; the thresholds were chosen on a second set, kept apart. "Known values" are those that can be derived from the data attached to each text, not every sensitive value in it. Precision on real text has not been measured yet: it needs a hand-labelled set, which is being prepared. ** Measured by us on 13 November 2026 with the PIIMB method (about 150,000 sentences in six tasks, at character level, model weights alone): the best of the other 36 models listed scores 0.836. Ours is first on average, not in every task, and the training splits of the same datasets are among its training data; the test splits were never used.

Next to the others

The same texts and the same measuring code for everyone. omissis-small has a tenth of the parameters of OpenAI's privacy-filter and, on the same graphics card, takes 0.5 ms per text against 32.5.

ModelParametersPIIMB, averagePIIMB, Italian sentencesReal texts: known values protectedTime per text*
omissis-small141 M0.9120.92798.8 %0.5 ms
omissis-base308 M0.9110.92598.8 %0.6 ms
OpenMed privacy-filter-multilingual-v21,400 M0.8360.87981.5 %16.7 ms
bardsai eu-pii-anonimization-multilang278 M0.6850.70588.3 %0.4 ms
openai/privacy-filter1,400 M0.6620.81179.0 %32.5 ms

* Averaged over the 150,000 benchmark sentences, on an RTX 4080 SUPER card, with our measuring code; on the CPU alone omissis-small takes about 26 ms for a short text. Measured on 13 November 2026. The PIIMB average is measured by us for Omissis and published by PIIMB for the others; the Italian sentences were measured by us for all of them. The real texts are a private set of 2,000 IT help-desk texts, the field Omissis is built for: the comparison is honest, not neutral. For Omissis the figure is that of the whole program (model plus rules); the omissis-base model alone reaches 95.1 %. We are not first everywhere: in three PIIMB tasks out of six another model does better, and on CVs bardsai is ahead of us when we keep every label on. All the numbers, their limits and the commands to redo the measurements.

What it recognises

People and contacts

people · e-mail · phone numbers · addresses · postcodes · towns · dates of birth

Identifiers

tax codes · VAT numbers · documents · number plates · VINs · case numbers · land-registry data · utility accounts

Payments

IBANs · cards · card security codes

Organisations

company names, including single-word ones

IT

passwords and keys · host names · IP and MAC addresses · domains · URLs · accounts · cloud identifiers · serial numbers

More, if you ask

dates · times · age · gender · titles · occupation · education · origin · places · health · religion · political views · orientation

Your choice

everything is on by default · if the AI has to work on exactly that data you can turn off payments, companies, IT data or case files · people, contacts and passwords can never be turned off

What is verified, and how

A privacy tool is judged by what it demonstrates, not by what it promises. This is where things stand today.

PartAutomated testsAgainst a real service
Terminal gateway33 checks, including tool-call arguments, signed reasoning, Codex on a subscription and a gateway of your ownYes: Claude Code on a subscription, a message and a file-writing action, with made-up data. Codex on a subscription not yet
MCP server for documents27 checks: no real value goes back to the assistant, limited folders, nothing with Omissis offYes: Claude Code read a made-up report seeing only placeholders. Inside Claude Desktop not yet
Browser extension40 checks in a real browser against a mock serviceNot yet
Desktop-app connector37 checks on a test appYes for ChatGPT: a message of made-up data reached the service as placeholders only. Claude Desktop not tested yet
Documents and scans21 checks: no original value in the output files, identical restoreNot applicable: it only works locally
Encrypted vault21 checks: no readable value in the file, stable labels after conversionNot applicable: it only works locally
Proxy9 formats against a mock serviceNot yet
Central console154 checks on the console and 42 on the PC side: rules accepted only by choice, per customer and per single PC, guided tour and guide inside the page, alerts, a hash-chained log, Microsoft sign-in against a mock serviceNot yet: a real Entra ID, a real Teams or Slack and a real mail server, and two different PCs on a network

What it does not do

For organisations and those who support them

Everything that runs on the PCs is free and open, in a company too and with no limit on the number of computers. Only the central console is paid, beyond 5 PCs or for several customers.

Rolling it out to many PCs

Scripts to install it silently and already configured, from an RMM, Intune or Group Policy, with the settings users must not change locked by the administrator. Free, in the repository.

One model for everyone

One PC lends its model to the others on the network: those with little memory neither load nor download it. Free, turned on from the panel.

Central console

All the PCs on one page: which ones have protection on and how much data stayed away from AI services, never the text itself. From there you send rules to the PCs that choose to accept them, per customer or per single PC, get alerts (protection off, a PC gone quiet, a new AI tool) and keep a compliance log in which any tampering shows. You install it on a Windows computer of your own. Free up to 5 PCs of one customer; with a licence beyond that, and for those who look after several customers.

Tailored work

Sector lists and formats (law firms, healthcare, tax, IT support), evidence for GDPR, NIS2 and AI Act compliance, help with the rollout.

The central console, with made-up data: connected PCs, alerts, data kept away from AI services
The central console, with made-up data: connected PCs, alerts, data kept away from AI services

Guide for organisations · The central console · Putting the console into service

Get in touch

For a console licence, a tailored solution, a partnership or a report:

francesco.gruner@mspincloud.com

Download

Code and model are public under the Apache 2.0 licence. Requires 64-bit Windows 10 or 11, 8 GB of memory (16 for the most accurate model) and 1 to 2 GB of disk; on an ordinary office PC choose the small model. The installer needs neither administrator rights nor Python: on first start it downloads the model, then it works offline and keeps itself up to date with signed updates. No GPU needed. The interface is in English and Italian; the guides are in both, the full reference of the program in Italian.