Skip to content
RankShark

RAGFlow

Open-source RAG engine with deep document understanding.

ToolOpen source · from $29/moby InfiniFlowsince 2024
71.3

SharkScore

#19 in Frameworks & SDKs

Why it scores 71.3

RAGFlow has a SharkScore of 71.3/100, ranked #19 of 665 in Frameworks & SDKs (updated 2026-11-18).

Each part compares RAGFlow with other listings in Frameworks & SDKs. Evidence covers 87% of the score's weight.

Adoption
55 · 30%
User reviews
no data · 25%
Maintenance
100 · 20%
Trust & transparency
80 · 15%
Momentum
no data · 10%

Paying never changes this score. How ranking works · method v2026-11-18.2

RankShark review of RAGFlow

Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review

We were not paid for this review.

As of 2026-11-18, RAGFlow is an open-source RAG and agent platform with self-hosted Docker deployment and paid cloud plans starting at $29/month. [E1][E2]

RAGFlow is an open-source retrieval-augmented generation engine and agent platform. It ingests and parses documents, retrieves relevant content, and supports chats and visual agent workflows. It can be deployed with Docker Compose or used through the listed cloud plans. [E1][E2]

Best for

  • Teams building document-grounded question answering with citations. · docs
  • Developers or operators who can deploy and maintain a Docker-based self-hosted system. · docs
  • Teams building visual agent workflows that combine retrieval, tools, and MCPs. · home page

Not for

  • Teams that need an API key on the Free plan; the pricing page says one is not available there. · home page
  • Teams that cannot provide an LLM and embedding model for a self-hosted deployment. · docs

Key features

  • Document ingestion and parsing. The vendor says its ingestion pipeline processes multiple data formats and structures them for retrieval. [E1] · home page
  • Hybrid search. The vendor says search combines vector search, BM25, custom scoring, and re-ranking. [E1] · home page
  • Document-grounded chat. The vendor's quickstart describes chats that answer from a dataset and can show citations to source content. [E2] · docs
  • Visual agent workflows. The vendor says its platform integrates RAG, tools, and MCPs in visual workflows. [E1] · home page
  • Self-hosted deployment. The vendor documents Docker Compose deployment and Docker images for Linux x86-64 and ARM64. [E2] · docs
  • Cloud plan limits. The pricing page lists app, team-member, dataset-storage, and monthly-credit limits for Free, Starter, and Pro. [E1] · home page

Pros

  • Self-hosted deployment is documented, with Docker images for Linux x86-64 and ARM64. · docs
  • Chats can show citations to the supporting document content. · docs
  • The vendor lists a Free plan and paid plans with published usage limits. · home page

Cons

  • The self-hosted quickstart recommends 4 CPU cores, 16 GB RAM, and 50 GB free disk space; actual needs depend on workload. · docs
  • Self-hosted setup requires access to an LLM and embedding model at runtime. · docs
  • The pricing page displays two amounts for Starter and Pro without explaining their difference. · home page

Pricing

PlanPriceLimits
Free$0 month5 Apps; 1 team members; 0.1 GB dataset storage; 500 credits / month; API key not available.
Starter$29 month; $59 month also displayed; the page does not explain the two amounts.50 Apps; 5 team members; 5 GB dataset storage; 5,000 credits / month; API key available.
Pro$129 month; $259 month also displayed; the page does not explain the two amounts.Unlimited Apps; 20 team members; 50 GB dataset storage; 20,000 credits / month; API key available.
EnterpriseContact usBYOC deployment; on-premises deployment; dedicated support; custom SLA.

Checked 18 Nov 2026

Free plan
Free plan: 5 Apps, 1 team members, 0.1 GB dataset storage, 500 credits / month, and no API key. · home page
Free trial
Not stated (checked 18 Nov 2026)
Trains on customer data
The vendor says it does not use Customer Data, imported documents, conversations, or prompts to train or fine-tune generalized AI or machine-learning models, and does not permit its model providers to do so. · privacy page
Data retention
The policy says content is retained until deletion or a deletion request; account and imported data are deleted within 30 days of a verified request, and encrypted backups expire within 90 days. · privacy page
Data regions
The vendor says personal data may be processed and stored outside the user's country, but does not name specific regions. · privacy page
Data processing agreement
The privacy policy says Customer Data processing is governed primarily by the customer contract and DPA. The policy references a Data Processing Agreement, but the fetched text does not provide its terms. · home page, privacy page
AI models used
The quickstart requires a Chat model and an Embedding model, configured through a model provider. It does not identify a specific required model. · docs
API
The pricing page says API keys are available on Starter and Pro, but not Free. The quickstart also documents an API health endpoint. · home page, docs
SDKs
The changelog says the Python SDK remains unaffected by the v1.0.0-rc1 changes. Other SDK availability details are not stated. · changelog
Webhooks
Not stated (checked 18 Nov 2026)
Platforms
The vendor documents Docker Compose deployment, with Docker images for Linux x86-64 and ARM64 starting with v1.0.0. · docs
Security claims
The vendor says it implements reasonable technical and organizational measures designed to protect personal data, while stating that it cannot guarantee absolute security. · privacy pageThe vendor says Google OAuth access and refresh tokens are stored encrypted. · privacy page

What we measured

GitHub stars · now
94,341
Latest release · now
v1.0.0 on 2026-11-06
Tranco traffic rank · list of the day
#532,631
How we tested · sources
  • Hourly uptime and response-time checks of the product's URL.
  • Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
  • We did not sign up, pay, or use the product as a customer.

Something wrong? Report an error with a public link that shows it; corrections re-run the review.

Top alternatives to RAGFlow

See all
  • 81.3

    Open protocol and SDKs that connect AI apps to tools and data.

    ToolOpen source
  • 77.7

    TypeScript toolkit for building AI apps and agents.

    ToolOpen source
  • 77.7

    Build agents and LLM apps in Python and TypeScript.

    ToolOpen source
  • 76.9

    Tracing, evaluation and monitoring for LLM apps.

    ToolFreemium · from $39/mo