Find the gap the literature is ignoring — with evidence any peer reviewer can reproduce.
You have documents. Gaptrium™ finds what's missing. Analyse hundreds of files in seconds — entirely in your browser. Zero upload.
Every AI platform imposes hard limits on file size and document count per session — typically 5 to 20 files, 10 to 25 MB each. This is not a pricing decision. It is an architectural constraint that cannot be removed no matter how much you pay.
Gaptrium™ processes documents locally in your browser. There is no server receiving your files — so there is no server imposing limits. The only ceiling is your device's processing power.
Any domain that produces documents has gaps in those documents. Gaptrium™ finds them — deterministically, reproducibly, and without sending your data anywhere.
Document and term limits depend on your plan. Zero capture applies on every plan — no document ever reaches our servers.
Every tool runs on the same engine — same corpus, same parameters, same result. Always.
Gaptrium™ analyses massive volumes of files and cross-references the information to find the pattern of coverage. It shows how themes, data, and arguments interconnect systematically — revealing hidden connections that would escape human reading.
Gaptrium™ isolates what the documentation omitted. The software scans the data corpus and pinpoints blind spots, intentional silences in reports, and information voids that require investigation.






AI is an extraordinary tool for many tasks. But when the problem requires deterministic results, reproducible findings, and zero data exposure — AI is structurally incapable of delivering. Not because of limitations that can be fixed with a better model or a higher-tier plan. Because probabilistic inference and deterministic analysis are fundamentally different things.
Gaptrium™ fills the gap AI leaves behind — the problems that require a finding to be reproducible by any reviewer, traceable to the source, and produced without your documents ever leaving your device.
These are not weaknesses that will be fixed in the next version. They are properties of how probabilistic models work — and why a different class of tool is needed for deterministic analysis.
| Capability | ✗ Why AI cannot do this | ✔ What Gaptrium™ makes possible |
|---|---|---|
| Document volume & file size | ✗ 5–20 files per session · 10–25 MB per file · loses context across documents · architectural constraint — cannot be fixed regardless of plan | ✔ No file size limit · Trial: 20 docs · Scope: 100 · Signal: 500 · Core: unlimited · bounded only by your device's processing power |
| Hallucinations & reliability | ✗ Invents facts, citations, clauses, and statistics that do not exist in the source documents. No way to detect or prevent — it is a property of probabilistic models | ✔ Zero hallucination — every output is a direct count from a specific document. Nothing is inferred, generated, or invented |
| Reproducibility | ✗ Probabilistic output changes on every run. Same question, different answer. No reviewer can reproduce or verify the finding | ✔ Byte-identical results every time. Same corpus + same parameters = same result — tomorrow, next year, or in peer review. Methodology published Springer Nature 2026 |
| Gap detection | ✗ Can summarise what is present but cannot map what is missing or systematically omitted | ✔ Specialised in auditing and isolating what was not written — the absence is the finding |
| Data privacy & compliance | ✗ Documents uploaded to cloud servers. Data retained for model training. Compliance requires contractual guarantees that vary by provider and jurisdiction | ✔ Zero upload — documents never leave your device. GDPR, UK GDPR & Swiss nFADP compliant by architecture. Safe for classified, legally privileged, and trade secret material |
| Pattern mapping | ✗ Loses global context in long readings. Cannot process the full corpus as a unified structure | ✔ Processes hundreds of files simultaneously as a single integrated corpus — full context, no memory loss |
| Auditability | ✗ Black box — opaque scoring, no traceability. Unusable for court submissions, regulatory filings, or audit sign-off | ✔ Every score traces back to a specific count in a specific document. Defensible in court, regulatory review, and peer-reviewed publication |
AI figures based on publicly documented limits as of 2026. Gaptrium™ is not positioned as a replacement for AI — it addresses a class of analytical problems that probabilistic models are structurally unable to solve.
Gaptrium™ is inspired by the MTTR methodology — published by the Intelligent Systems Research Centre, London Metropolitan University, in partnership with CTI Renato Archer (Brazilian Federal Government) — and independently validated across four institutional contexts: ITNG / Springer (2020), IEEE Transactions on Services Computing, ECKM · Coventry UK (2021), and Springer Nature Systematic Reviews (2026). Gaptrium™ shares no code, screen or implementation with MTTR.
See the full methodology, genealogy & institutional citations →Not an AI platform — a Deterministic Intelligence Platform. Same corpus, same parameters, same result. Always. Reproducible. Auditable. Zero retention.
For: Academic Research · Media Intelligence · PR & Narrative Analysis · Commercial Due Diligence · Government & Audit
Documents are processed entirely in your browser on every plan. No file reaches our servers. Compatible with classified information, legally privileged material, and trade secrets.