TTalk 言几 · The Talk

Software and data,
built around large language models

We are a software and data company built around large language models. We turn models into handy products — Guanlan reads your material, Jinshu writes long fiction — and build the data layer beneath them: libbitcask packs KV, full-text, and vector search into one embedded engine. The agent and web frameworks we built along the way are open sourced too.

What we do

Three lines, one core

The large language model is the core. Software puts it in people's hands, data makes it work well, and frameworks let developers build fast.

Software products

Desktop apps that put large language models in ordinary people's hands: organizing reading material and writing long fiction. Everything lives as ordinary files on your own computer.

  • AI research assistant
  • AI long-form writing

Data products

LLM applications live on retrieval: full text, vectors, hybrid fusion, plus transactions and graphs. We write our own storage and retrieval engines and embed them in the app — no extra service to run.

  • Embedded KV + search engine
  • BM25 / vector / hybrid search
  • Erlang/OTP data layer

Frameworks

The frameworks we built for our own products, cleaned up and open sourced: AI agent frameworks for Erlang/OTP and Clojure, template and Markdown rendering, and a server-rendered web framework.

  • Erlang/OTP agent framework
  • Clojure agent framework
  • Rendering and web frameworks

Software products

Three handy tools

Write code, organize reading, write novels. Installed on your own computer — your files stay yours.

Software product

Guanlan

Watch your knowledge flow

An AI research assistant that lives on your computer. Point it at a folder to ask questions, organize cards, draw link graphs, build slides.

Software product

Jinshu

The AI writing partner that remembers your world-building

An AI partner for long-form fiction. Characters, foreshadowing, plotlines, what the reader already knows — all remembered, with self-checks after every chapter.

Software product

Coxswain

An Emacs-style code editor, rebuilt in Clojure

An Emacs-style editor rebuilt in Clojure: Emacs keybindings, tree-sitter highlighting, LSP, project tree, Git, REPL, and themes. Free download.

Data products

The data layer for LLM applications

Search, transactions, and graphs in one embedded engine. Guanlan's search runs on it.

Frameworks

Tools for building LLM applications

AI agent frameworks for Erlang/OTP and Clojure, plus a rendering engine and a web framework — all Apache 2.0 open source.

Framework

BeamAI Extra

Tools, RAG, and protocol extensions for BeamAI

Extensions on the BeamAI core: file / shell / human-interaction tools and middleware, RAG, an MCP client and server, the A2A protocol, and an AG-UI event-stream server.

Framework

BeamAI Render

Template and Markdown rendering for Erlang

Mustache and Jinja2 templates compile to Erlang modules at build time — rendering is one function call — plus a Markdown engine ported from markdig with 30 extensions. Zero dependencies.

Framework

aihtml

Web pages written in Erlang functions

Assemble pages from Erlang functions with complete server-side HTML; clicks are answered statelessly by Erlang functions and server pushes ride SSE. 112 components, four theme axes, SEO-friendly.

Framework

clj-agent

The AI agent framework for Clojure

Build AI agents in Clojure: deftool for declaring tools, ChatClient and filter middleware, five hook types, a SimpleAgent that can pause for approval, and adapters for a dozen-plus model vendors.

How we work

AI is here to help, not to take over

  • Your files belong to you

    Your material and drafts are ordinary files on your computer — open them in any other program, back them up whenever you like. Guanlan never modifies a single original file.

  • Runs on your own computer

    No account required; searching your material happens locally. AI conversations can use DeepSeek, Qwen, Kimi and other online services with your own account.

  • Changes pass through you

    AI never edits quietly. Jinshu marks every revision word by word and applies it only after you confirm; Guanlan puts its output in a draft folder until you approve it into the library.

  • We build our own foundations

    Guanlan's search runs on our own data engine, libbitcask. No external database service — install and it just works.

Where the name comes from

言(yán)几(jī) · The Talk

夫易,圣人之所以极深而研几也。

—— Book of Changes, Xici I (classical Chinese)

Together: seeing the sign in language, seizing the moment in the subtle. Yán is the software that lets models converse; jī is the craft of pulling signal out of data and retrieval.

Read the full story →

Explore our products