Perimattic
ManageStacks · AI & ML

Managed RAGFlow Hosting — production-ready from $55 a month

Retrieval-augmented generation engine for building document-grounded AI assistants. Deployed on your own dedicated instance in AWS, Azure, or GCP, kept patched, backed up, and monitored by ManageStacks — standard RAGFlow, no lock-in.

RAGFlow on ManageStacks is the open-source retrieval-augmented generation engine deployed to your own AWS, Azure, or GCP region, with pricing sized to the instance it needs. Teams use it to turn documents into a knowledge base that grounds AI assistants and chatbots in their own content. It is a self-hosted alternative to hosted RAG services and chatbot platforms such as Azure AI Search with OpenAI.

Daily backups includedAWS · Azure · GCPData export any time24×7 SRE available
RAGFlow logo
RAGFlow
Retrieval-augmented generation engine for building document-grounded AI assistants
The application

What does RAGFlow do, and why do teams deploy it?

RAGFlow is an open-source engine for building generative AI applications on top of your own documents, using retrieval-augmented generation. You upload files, the system parses and chunks them, stores them for retrieval, and then lets a language model answer questions using the retrieved passages, so responses can be traced to source material. The project is written in Python with a TypeScript front end and runs several supporting services, including a document store and search index, which is why it needs a larger instance than most applications. It is used to build assistants for internal knowledge, support, and document question answering.

  • Document ingestion with parsing and chunking for retrieval
  • Knowledge bases that ground model answers in your files
  • Chat assistants built over selected knowledge bases
  • Connections to language models you choose
  • Answers that reference the source passages they used
  • Web interface for managing documents and assistants
Neural network visualization representing machine learning inference
AI & ML

Retrieval-augmented generation engine for building document-grounded AI assistants

Open-source alternative

RAGFlow as a self-hosted alternative to Azure AI Search and Amazon Kendra

Instead of paying per-seat fees to a SaaS vendor, teams self-host RAGFlow to keep full control of their data and infrastructure. ManageStacks removes the operational overhead — you get the cost and privacy benefits of open source with the reliability of a managed platform.

Replaces Azure AI SearchReplaces Amazon Kendra
Pricing

What does managed RAGFlow hosting cost?

Flat per-app pricing, in your chosen AWS, Azure, or GCP region. No per-user pricing — a busy deployment costs the same as a quiet one.

Starter

$55/app/mo

Staging and internal tools. Dedicated instance, TLS, daily backups, managed upgrades.

Standard

$69/app/mo

Production workloads. Adds monitoring, staging environment, region choice, priority support.

Business

$89/app/mo

High-traffic and compliance workloads. Adds a high-availability replica and same-day support.

24×7 SRE retainer

$499/mo

Round-the-clock on-call across every hosted application, for teams that need a pager answered at 3am.

The honest answer

When is self-hosting RAGFlow the right answer instead?

“Managed hosting is not always the correct call.”

Self-host when a platform team already runs the infrastructure and on-call rotation to operate RAGFlow at genuinely low marginal cost. Self-host when compliance requires an air-gapped or on-premises deployment that no hosted option can satisfy. And self-host when the deployment depends on heavy customisation with a fast internal build-deploy loop, because an internal release process will beat any managed change process.

For everyone else — teams whose engineers have better things to do than shepherd upgrades — managed hosting is cheaper than the hours it replaces.

Infrastructure

Which cloud should RAGFlow run on — AWS, Azure or GCP?

For most workloads, the choice of cloud matters less than proximity: run RAGFlow in the same cloud and region as the applications and data it talks to, because every request between them adds a round trip. The underlying compute performs equivalently across AWS, Azure, and GCP.

In practice, an existing cloud footprint decides it. All plans support all three clouds, and moving regions later is a scheduled migration, not a rebuild.

AWS logo
AWS

Deepest managed-service catalog, default when there's no existing footprint

Azure logo
Azure

Best fit for teams already on Microsoft 365 or Entra ID

GCP logo
GCP

Strongest for data/analytics-adjacent workloads

GPU compute hardware used for AI model inference
Multi-cloud

Every plan supports AWS, Azure, and GCP — region choice included.

FAQ

Common questions about RAGFlow on ManageStacks

What is RAGFlow on ManageStacks?

RAGFlow is an open-source engine for retrieval-augmented generation, helping AI assistants answer from your own documents. ManageStacks deploys and operates it in your AWS, Azure, or GCP account, so your documents and indexes stay in infrastructure you control.

How does RAGFlow compare to hosted chatbot platforms?

Hosted platforms handle ingestion and chat on the vendor's servers, so your documents leave your environment. RAGFlow is self-hosted, keeping files and the retrieval index in your cloud, and you decide which language model it talks to.

What does retrieval-augmented generation mean in practice?

Instead of relying only on what a language model memorized, the system first finds relevant passages in your documents and gives them to the model alongside the question. Answers are therefore based on your content and can point back to their sources.

Do I keep control of my documents and knowledge bases?

Yes. Uploaded files and the indexes built from them are stored in your own cloud account. You can remove documents or delete a knowledge base at any time, and nothing is held by a third-party service.

Why does RAGFlow need a larger instance?

It runs several components together, such as the application, a document store, and a search index, and document parsing is resource-hungry. ManageStacks sizes the instance for those needs, which is why pricing depends on the instance rather than a fixed per-app figure.

Run RAGFlow without carrying the pager

Subscribe through your AWS, Azure, or GCP marketplace. We handle provisioning, SSL, monitoring, backups, updates, and security. From $55/mo.