Document processing
Test formats, layouts, tables, OCR and chunking.
WZ-IT designs and operates RAGFlow for demanding knowledge repositories and aligns parsing, embeddings, retrieval, models and infrastructure with the use case.
The following are trademarks of their respective owners: RAGFlow (InfiniFlow). WZ-IT is an independent service provider and has no business, partnership, or contractual relationship with these companies. We offer independent migration, installation, hosting, and operations services.
With RAGFlow, file formats, layouts, tables, chunking, embeddings and search strategy determine whether users receive reliable sources. Merely starting the software does not answer these questions.
We test representative documents, define quality questions and then design components, storage, model endpoints and operating processes.
Production setups include multiple data, search and background services. Resource requirements and operations depend on document volume, ingestion load, concurrency and models.
The calculator covers standard operations. Parsers, data sources, models and custom RAG integrations are scoped separately in the proposal.
Parsing, chunking, embeddings, retrieval and answer models must be evaluated with real documents rather than treated as a simple container stack.
Test formats, layouts, tables, OCR and chunking.
Operate vectors, keyword search, metadata and reranking.
Select embedding, reranking and answer models for language and quality.
Monitor ingestion, failures, storage, backup and upgrades.
WZ-IT operates the platform and pipeline; source approval and business validation are shared with the customer.
| Area | Responsibility | Scope and boundaries |
|---|---|---|
| RAGFlow platform | WZ-IT | Deployment, data and search services, monitoring and updates. |
| Document pipeline | WZ-IT | Technical parsing, chunking, embedding and retrieval. |
| Test corpus | Shared | We create measurements; the customer provides representative material. |
| Permissions and sources | Shared | Access controls and data paths are designed per integration. |
| Business approval | Customer | The customer validates correctness and permitted use. |
| Custom connectors | Optional WZ-IT service | Automated sources and ACL sync are developed separately. |
Prepare PDFs and other formats for search with suitable parsing and chunking methods.
Configure vector, keyword and hybrid search methods for the content and questions.
Integrate approved document sources through available interfaces or custom pipelines.
Assess retrieval results, sources and answers with a representative test set.
Protect users, administrative interfaces, model access and data paths for the target architecture.
Embed RAG capabilities into internal assistants and business applications through defined interfaces.
Assess existing data and configuration, migrate them in a test run and move to managed operations through a controlled cutover.
Secure SSO, roles, administrative paths and external access for the application and existing infrastructure.
Back up all stateful components consistently and document the recovery path for the agreed scope.
Monitor and update the application and its technical dependencies and operate them under the agreed service level.
A clearly defined operating scope instead of an opaque hosting flat fee.
We set up a test instance for you, usually on the next business day. No payment details required. After seven days it is deleted unless you continue.
We combine the right compute size with ongoing operations, backups, monitoring and a service level appropriate for the criticality of RAGFlow. High availability and recovery targets are designed separately where needed.
We also design custom hosting architectures, integrations and migrations around RAGFlow. Contact us for a technical assessment.
One managed standard RAGFlow application is included in the Starter workload. Business and higher levels add a flexible operations allowance for planned work during regular service hours. Select compute, additional applications, storage and the appropriate service level.
A workload is one compute instance with the applications agreed for it.
One standard app per workload is already included. Additional dedicated servers count as separate workloads.
€79.90 per started TB and month, including daily encrypted offsite backup with 7-day retention.
Enquiry
Briefly describe the current state and objective for RAGFlow. We assess infrastructure, integration, and ongoing operations.
Page count is insufficient; OCR, tables, update rate, context and concurrency determine resources.
| Usage scenario | Technical starting point | Key factors |
|---|---|---|
| Pilot with selected documents | Test corpus and evaluation | Parser, chunking and search are tuned first. |
| Large heterogeneous corpus | Ingestion and storage assessment | OCR, workers and rebuild times are measured. |
| Many concurrent users | Retrieval and model load test | Search and inference are sized separately. |
| Critical knowledge system | Recovery and quality design | Backup, reindexing and evaluation are planned. |
Document profiles, volume, models and quality target are tested before a proposal.
Sources, search and models are placed according to protection and performance needs.
European operation after data and model assessment.
Deployment close to object and data services.
Documents, search and models inside your network.
Local data with controlled model or search services.
Sources, processing, index, models and user access remain distinct and auditable.
Search, chat and integrated RAG APIs.
TLS, identity, roles and defined APIs.
Permissions are planned per collection and integration.
Parsing, chunking, retrieval, reranking and answers.
Originals, parsing results and protected artefacts.
Chunks, embeddings, metadata and search state.
Parsing, embedding, reranking and answer services.
RAG does not automatically eliminate hallucinations; quality requires repeatable tests with known sources.
Answers about documents, models, permissions and quality.
It is relevant for document-centred RAG. We test real PDFs, scans, tables and business questions.
Yes, when documents, search and models are deployed on suitable local resources.
Collections, groups and source rights are designed per integration; complex ACL sync is separate work.
Document complexity, OCR, indexes, models and quality objectives materially change the architecture.
Yes. We compare parsing, chunking, retrieval and models against an agreed corpus.
As an alternative to managed hosting in the data centre, WZ-IT provides the hardware, configures RAGFlow, and handles hardening, monitoring, updates, backup and technical support. Access can be limited to the internal network or enabled through VPN and existing identities.
from EUR 349 excl. VAT / month · plus one-time provisioning and initial setup

08.11.2025
The use of Large Language Models (LLMs) such as GPT-4, Claude or Llama has evolved from experimental applications to mission-critical tools in recent years. However,...
09.11.2025
Local AI inference means running a language or multimodal model on owned hardware rather than through a public API. Organisations gain control over data paths,...
24.11.2025
OpenAI released GPT-OSS 120B as an open-weight reasoning model on 5 August 2025. Its native MXFP4 quantisation allows OpenAI to position the model for a...
These solutions are often used together with RAGFlow
These solutions offer similar functionalities and can be evaluated together
These solutions are direct alternatives with similar use cases
No risk: worst case, you leave with a clearer understanding of your project than before.


“WZ-IT's advice on our Azure migration was technically sound and completely non-binding right from the intro call - we took away a great deal.”
Whether a specific IT challenge or just an idea - we look forward to the exchange. In a brief conversation, we'll evaluate together if and how your project fits with WZ-IT.