EXTENDED COURSE

AnythingLLM

Five hands-on lessons turn AnythingLLM from a fresh install into a private ChatGPT over your own documents. You install the free desktop app, point it at a local model so nothing leaves your machine, drop in your PDFs, and learn to get answers with in-chat citations. From there you organise work into clean workspaces, give the assistant agents and Community Hub skills, and finally self-host the Docker server and its developer API for a whole lab. Where Ollama and LM Studio run models, AnythingLLM adds document-RAG and agents on top — a private assistant over your own knowledge.

1Lessons5step-by-step, ~90 min each

Setup

00 Install AnythingLLM & chat with your first PDFs

Basics

Advanced

2Cheat sheetcopy-ready expressions

3Examples3what people built

Internal tools & ops1

AnythingLLM Small biz Founder

Course starter

Support replies grounded in your product docs

A small team ingests product manuals, FAQs, and a sample of resolved tickets into an AnythingLLM workspace running on a local model. When a new question comes in, support pastes it and gets a draft reply grounded in the actual docs, with citations to check before sending — customer data and internal docs stay on the machine.

First-draft support answers in seconds that are anchored to real documentation, with no per-token cost and no customer data sent to a cloud API.

Try it yourself

In AnythingLLM, create a "Support" workspace on a local model, upload your manuals + FAQ + a few resolved tickets, then ask: "Draft a reply to this customer question using only our docs, and cite the source: <paste question>."

Research & data tools1

AnythingLLM Scientist

Course starter

Private literature review over your own PDFs

A researcher drops a folder of papers — including unpublished manuscripts and licensed PDFs — into an AnythingLLM workspace and points it at a local model (via Ollama or LM Studio). They then ask cross-paper questions and get answers with citations back to the exact source document, with nothing ever leaving the laptop.

A grounded, citeable assistant over sensitive or licensed papers, with zero risk of the PDFs reaching a cloud provider or training set.

Try it yourself

Install AnythingLLM, set the LLM provider to a local model (Ollama / LM Studio), create a workspace, drag in your PDFs, then ask: "Across these papers, what methods were used to measure X, and which reported effect sizes? Cite each source."

Knowledge & docs1

AnythingLLM AI-native SME Founder

In the gallery

Private team knowledge base from your own docs

IT self-hosts the AnythingLLM Docker server on the office network and ingests the staff handbook, SOPs, and internal wiki into a shared workspace. Employees chat with it like an internal ChatGPT, and every answer links back to the source document — all running against a local model so proprietary content stays in-house.

Staff get instant, cited answers from company docs without a SaaS chatbot ever holding the proprietary text on external servers.

Try it yourself

Deploy AnythingLLM via Docker, set a local model provider, create a workspace, upload your handbook / SOP / wiki exports, and share the URL internally. Ask: "What is our parental-leave policy and who approves it? Cite the policy."