AnythingLLM vs RAGFlow
Side-by-side comparison of AnythingLLM and RAGFlow for local, private document Q&A.
| AnythingLLM | RAGFlow | |
|---|---|---|
| License | MIT | Apache-2.0 |
| Latest release | v1.18.0 | v1.0.0-rc1 |
| Release date | 2026-10-13 | 2026-09-29 |
| Hardware | Desktop or Docker; built-in engine runs CPU or GPU (NVIDIA CUDA, AMD/Intel Vulkan) | 4 cores, 16GB RAM, 50GB disk recommended (Docker) |
| Document types | PDF, DOCX, TXT, CSV, web pages, audio | PDF, scans, DOCX, XLSX, PPT, images |
| OCR / tables / retrieval | Basic text extraction; built-in reranker unverified | Strong: layout-aware parsing, OCR, table recognition, citations unverified |
| Cloud calls | None by default; optional cloud LLM/embedder providers | None if pointed at Ollama/local embedder unverified |
| Fully offline | Yes | Yes |
| Status | Active | Active |
Last updated 2026-10-24. Fields marked unverified are summarized from docs and reviews and have not been checked against primary sources yet.
The RAGShelf weekly
New local RAG tools, releases, and tested setups. One email a week, no spam.