Integrated with AI

Cloud and on-premise AI powered by leading Large Language Models

ChronoScan AI — Next-Level Document Intelligence

ChronoScan Advanced and Enterprise editions support the latest generation of AI, running Large Language Models (LLMs) for transformative document automation. LLMs deliver advanced data extraction, document analysis, automatic content classification, summarization, natural language understanding, and more.

Drastically reduce manual effort on complex and unstructured documents
Advanced content classification, entity extraction, and summarization out of the box
Choose cloud AI (maximum performance) or local models (total data privacy)

Cloud AI providers

OpenAI
OpenAI GPT / Azure OpenAI
ChatGPT 3.5, 4, GPT-4o, GPT-5
  • Integration with ChatGPT 3.5, 4, GPT-4o and GPT-5 online services
  • Context windows up to 1M tokens (4K, 16K, 32K, 128K supported)
  • GPT-4-vision and GPT-4o support direct image processing — no OCR needed
  • Highly flexible for extracting, classifying, or summarizing any data
  • Scriptable via VBScript for direct AI-powered automation
No OCR required, faster for image requests
Consumes credits, requires internet, images processed on OpenAI / Azure servers
Gemini
Gemini AI (Google)
Gemini via Google AI Studio or Vertex AI
  • Enterprise-grade text and image analysis, semantic understanding, and entity extraction
  • Multilingual support, entity recognition, and data classification
  • Secure and scalable with Google Cloud infrastructure
  • Handles images, tables, and formatted text natively
  • Data residency: run Gemini on Vertex AI and choose the processing region (EU, US or global). Documentation NEW
No OCR required, faster for image requests
Vertex AI: documents processed inside the region you select, under Google Cloud enterprise terms
Consumes credits, requires internet, images processed on Google servers (AI Studio or your chosen Vertex AI region)

Data residency with Gemini on Vertex AI

Choose where your documents are processed NEW

Gemini requests can now run on Google Vertex AI in the region you select, instead of the default Google AI Studio endpoint. Same models, same credits, same job configuration — only the location of the processing changes. Ideal for GDPR, public sector and regulated industries.

EU
or US / global
1 Select the endpoint

In the Gemini configuration, switch from Google AI Studio to Vertex AI. ChronoScan Hub issues short-lived access tokens — no Google Cloud account or service keys on the workstation.

2 Pick the region

Choose EU, US or global per workstation, or let your administrator enforce a region centrally for the whole license from ChronoScan Hub. Available regions per model are published automatically.

3 Keep working as usual

Documents are sent directly from ChronoScan to Vertex AI in the selected region. Results, fields and credits behave exactly as before, and every request is logged with its endpoint and region for auditing.

Processing stays inside the selected region
Google Cloud enterprise terms & compliance
Region enforceable per license by the administrator
Endpoint & region visible in usage reports

Available from ChronoScan 1.0.4.15 with ChronoScan Hub. Vertex AI requests are processed in real time; no extra cost over the standard Gemini tariff.

Setup guide: Gemini on Vertex AI

Asynchronous batch processing

AI Batch Processing — Deferred Mode NEW

Submit large document volumes to OpenAI asynchronously via ChronoScan Hub — results delivered back automatically, at a significantly lower cost.

40%
cost saving
1 Submit

ChronoScan groups your documents into a batch and submits it to OpenAI's Batch API via ChronoScan Hub — no changes to your job configuration needed.

2 Process

OpenAI processes the batch asynchronously — typically within a few hours. ChronoScan Hub monitors progress and manages the queue automatically.

3 Deliver

Results are returned to ChronoScan automatically. Fields are populated exactly as with standard AI requests — fully integrated into your existing workflow.

Up to 40% lower AI cost vs real-time
No job reconfiguration required
Hub monitors & delivers results automatically
Ideal for overnight or high-volume runs

Requires ChronoScan Hub with an active account. Enabled per customer — contact us to activate for your account.

Documentation

On-premise / Private AI

Meta Ollama Mistral
Open-Source LLMs — Llama, Mistral, Gemma & more
Via Ollama or compatible platforms
  • Supports Llama (Meta), Mistral, Gemma and more via Ollama or similar platforms
  • Run locally or on your private server — full data confidentiality, zero cloud dependency
  • Ideal for strict privacy or compliance requirements
  • Low-cost way to experiment with LLM technology in-house
Technical note

Each LLM has a context window (max tokens at once): 4K, 16K, 32K, 128K… More tokens = richer analysis but higher compute. GPU recommended for large-scale jobs.

No internet required, complete data privacy, no per-query costs
GPU recommended, performance depends on local hardware

Getting started guides

Ollama & GPT Quickstart
PDF Guide — V1
Setting up a Job with AI
Configuration PDF