ToolsRecipesCompareCategoriesHardwareDGX SparkModelsMCPLocal vs cloudCost calculatorPeopleQuestionsSubmitAboutSponsorNewsletterFollow on X

โ† All tools

FastGPT

New Active Docs only Last checked 2026-11-16 Custom / mixed Server / Docker

Website ยท Source

๐Ÿ“ฌ One weekly email: local RAG releases, hardware compatibility updates, and practical setup guides. Free. Unsubscribe anytime.

We use your email only for the newsletter. Privacy policy

Updates on X: @mkjerman

What it is

FastGPT is a self-hosted knowledge-base and agent platform from Labring (the Sealos team). You upload documents into knowledge bases, then build chat apps or visual workflows on top, with hybrid search, reranking and citations you can review and edit.

It runs with Docker Compose. Its official docs show how to connect local LLMs and embedding models through Ollama. The license is Apache-2.0 with extra conditions (no competing multi-tenant SaaS, keep the logo), so read it before building a product on it. The README is mostly in Chinese, with an English version. RAGShelf has not run it.

License
FastGPT Open Source License (Apache-2.0 plus conditions: no multi-tenant SaaS, keep logo)
Latest release
v4.17.4
Release date
2026-11-05
Hardware
Docker Compose server (install script + docker compose up); web UI on port 3000 unverified
Document types
TXT, MD, HTML, PDF, DOCX, PPTX, CSV, XLSX, URLs, CSV bulk import unverified
OCR / tables / retrieval
Hybrid search + reranking; citations shown in chat and editable; visual workflow builder unverified
Cloud calls
Models are configured in its admin (AI Proxy); official docs show local LLMs and embeddings through Ollama. A hosted cloud version also exists unverified
Fully offline
Yes
GitHub stars
30,043
Last commit
2026-11-16 (last push, GitHub API 2026-11-16)

Install

Copied from the official README. Check the project docs for options.

bash <(curl -fsSL https://doc.fastgpt.io/deploy/install.sh)
docker compose up -d

Retrieval

Hybrid (vector + keyword)

Based on: README: hybrid retrieval and reranking over knowledge bases (checked 2026-11-16) (checked 2026-11-16). Vector vs. vectorless RAG explained โ†’

NVIDIA DGX Spark

DGX Spark: Unknown

No DGX Spark report found yet. We only mark a tool as working when there is a published source.

DGX Spark guide โ†’

Last updated: 2026-11-16. License and release data from the GitHub API. Fields marked unverified have not yet been checked against official docs. How we know: Docs only (see the legend). Spot something wrong? Send a correction.

๐Ÿ“ฌ Get next week's tool

One new local RAG tool in your inbox every Monday. Free, unsubscribe anytime.

Sample issue: "AirLLM: run a 70B model on a 4GB GPU by loading one layer at a time. What it's good for (slow, private batch Q&A over your docs), what it isn't (fast chat), and the install command." Plus one new setup recipe or page, like our local vs. cloud cost calculator.

We use your email only for the newsletter. Privacy policy