Supported Formats

100+ File Formats
Fully Supported

Upload your original documents as-is and start AI analysis instantly — no conversion required.

PDF

Adobe PDF

HWP

HWP/HWPX

Hangul Word

DOC

DOCX/DOC

MS Word

PPT

PPTX/PPT

MS PowerPoint

XLS

XLSX/XLS

MS Excel

TXT

TXT/CSV

Text Files

XML

XML/JSON

Structured Data

HTM

HTML/MHT

Web Pages

RTF

RTF/ODF

Compatible Formats

IMG

JPG/PNG

Images (OCR)

EML

EML/MSG

Email

+

90+

More Formats

Core Features

Core Features of Meerae AI

AI handles every task on your documents — search, summarize, answer, and cross-reference.

Semantic Intelligent Search

Find accurate information through meaning-based search rather than keyword matching. The engine understands synonyms, paraphrases, and context — searching for "revenue target" also surfaces "income plan" and "sales outlook."

Semantic Search Context Understanding Similarity Matching Multilingual

Search comparison: "this year's performance outlook"

Keyword Search

Results: 0 — no exact match for "this year's performance outlook"

Meerae AI Semantic Search

98% "2026 Revenue target is KRW 15.0B..."

94% "This year's income plan is 35% YoY..."

91% "H2 sales outlook revised upward..."

56-page audit report → 3-line summary

PDF

audit_report_2025.pdf 56 pages

Analysis complete

AI Summary

1. FY2025 total Revenue KRW 11.2B, +28% YoY, operating margin 15.3%.
2. AI Solutions division grew to 45% of Revenue, +67% YoY.
3. Healthy balance sheet: debt ratio 42%, cash on hand KRW 3.2B.

AI Automatic Summarization

Boil down dozens of pages into a 3-line executive summary. Meerae AI understands document structure, weighs importance, and extracts only what matters for decision-making.

3-Line Summary Key Point Extraction Structure-Aware

Natural Language Q&A

Chat with your documents. Ask questions in plain language — no special search syntax — and receive accurate, source-cited answers grounded in the document content.

Chat Interface Source Citations Follow-up Questions

Conversation example

What's the penalty clause in the contract?

Upon termination, the penalty is 20% of the remaining contract value, waived under force majeure.

📄 contract.docx, Art. 15 §3

What counts as force majeure?

Natural disasters, war, regulatory changes, and pandemics are covered.

📄 contract.docx, Art. 18 §1

Multi-document Cross-reference

PDF

Business Plan

HWP

Proposal

XLS

Financials

Cross-analysis result

⚠ Revenue target mismatch between Business Plan and Proposal

✓ Cost line items match between Financial Statements and Business Plan

⚠ Headcount plan: Business Plan 50 vs. Proposal 45

Multi-document Cross-reference

Analyze multiple documents at once to verify information consistency automatically. Meerae AI flags numerical discrepancies and inconsistent statements across files and reports them.

Cross-validation Inconsistency Detection Unified Analysis

Core Tech / Architecture

RAG-Powered
Intelligent Answering

Retrieval-Augmented Generation (RAG) architecture delivers accurate, document-grounded answers you can trust.

1

Document Upload

Auto-recognize 100+ formats and extract text

2

Chunking & Embedding

Split into semantic chunks, convert to vector embeddings

3

Vector DB Storage

High-dimensional vector indexing for similarity search

4

Context Retrieval

Retrieve Top-K chunks most relevant to the query

5

LLM Answer Generation

Retrieved chunks + LLM produce source-grounded answers

Semantic vs. Keyword Search

Method

String Match

Semantic Similarity

Synonyms

Not Supported

Auto-detected

Context

None

Fully Supported

Accuracy

~45%

92%+

Keyword Search

Meerae AI

Vector Database Architecture

Document → Chunk Split

Chunk #1

Chunk #2

Chunk #3

...

Chunk #N

↓ Embedding Model

Vector Space (High-Dimensional)

dim 1

0.82

dim 2

-0.15

dim 3

0.67

dim N

0.41

↓ Cosine Similarity

Top-K Similar Chunks Returned

Chunk #47 — 98.2%

Chunk #12 — 94.7%

Chunk #83 — 91.3%

Security & Privacy

Enterprise-Grade
Security Architecture

A multi-layered security framework designed to handle sensitive documents safely.

Data Encryption

All data is encrypted in transit (TLS 1.3) and at rest (AES-256). Both original documents and vector data remain encrypted at all times.

✓TLS 1.3 in-transit encryption

✓AES-256 at-rest encryption

✓Key Management System (KMS)

On-Premise Deployment

Install and run Meerae AI directly on customer-side servers. No data ever leaves your network, ensuring the highest level of security.

✓Direct install on internal servers

✓No external data transmission

✓Docker / Kubernetes support

Access Control

Role-based access control (RBAC) provides fine-grained permissions per document and per user. Comprehensive audit logs track every access event.

✓RBAC role-based permissions

✓Per-document access settings

✓Complete audit logging

Use Cases

Deployed Across
Industries

Delivering transformative efficiency in document-heavy industries — legal, finance, government, and enterprise.

Legal / Law Firms

Case law & contract analysis

Instantly search relevant clauses across thousands of cases and contracts, and automatically surface similar precedents for comparison. Cuts legal research time by up to 80%.

Example use cases

• Risk analysis per contract clause
• Automated similar-precedent search and comparison
• Monitoring of regulatory changes

Finance / Investment

Report & disclosure analysis

Rapidly analyze corporate filings, audit reports, and investment memos. Automatically extract key financial indicators and risk factors to dramatically accelerate investment decisions.

Example use cases

• Automated extraction of audit report KPIs
• Multi-year Financial Statements comparison
• Automatic identification of investment risk factors

Government / Public Sector

Policy & citizen-request document analysis

Instantly search across vast volumes of policy documents, legislation, and citizen requests. Policymakers can quickly identify relevant precedents and legal grounds.

Example use cases

• Consistency checks across policy documents
• Search for similar citizen-request resolutions
• Impact analysis of legislative changes

Enterprise / Consulting

Internal document & report management

Unify and instantly search internal policies, process documents, and project reports. From new-hire onboarding to project insight reuse — maximize productivity across the organization.

Example use cases

• Instant search of internal policies and guidelines
• Reuse of insights from past projects
• Support for new-employee onboarding

Key Numbers

Meerae AI performance, by the numbers.

0

or more

File Formats Supported

3s

or less

Average Response Time

0

% or higher

Answer Accuracy

0

MB per file

Maximum File Size

Frequently Asked Questions

Common questions about Meerae AI.

Which document formats are supported?

Meerae AI supports 100+ formats including PDF, HWP, HWPX, DOCX, DOC, PPTX, PPT, XLSX, XLS, TXT, CSV, XML, JSON, HTML, RTF, ODF, EML, and MSG. Image files (JPG, PNG) are processed via OCR to extract text automatically for analysis.

How is answer accuracy ensured?

We use a RAG (Retrieval-Augmented Generation) architecture so every answer is grounded in the source documents. Each response includes citations (document name and page), and hallucination-mitigation techniques prevent the model from inventing information that isn't in the documents.

Is on-premise deployment available?

Yes. We support both cloud (SaaS) and on-premise deployments. For organizations with strict security requirements, Meerae AI can be installed on customer-side servers so that no data ever leaves your network. Docker and Kubernetes environments are supported.

Can multiple documents be analyzed together?

Yes — multi-document cross-referencing is built in. Upload several documents at once and Meerae AI will cross-analyze them, automatically linking related information and flagging numerical or contextual inconsistencies. Questions like "Compare the Revenue targets in the business plan and proposal" work out of the box.

Can it integrate with existing systems?

Yes. A REST API enables straightforward integration with existing business systems (ERP, groupware, document management, and more). Single Sign-On (SSO) is supported so you can reuse your existing identity provider without separate account management.

What's the pricing and rollout process?

Pricing is customized based on usage scale and deployment model. A typical rollout proceeds as (1) requirements analysis, (2) PoC, (3) customization, and (4) deployment and training. The PoC is usually completed within about two weeks so you can validate value quickly.

Transform Your Document Workflow
Starting Today

With Meerae AI, find what you need across thousands of pages in seconds
and get instant summaries of the essentials. Talk to our experts for the right deployment plan.

Get in Touch

meerae.info@gmail.com