ThoughtDAG

Turn branching LLM conversations into a visible, editable thought graph.

#canvas #graph #open-source #local-first

Quick answer

ThoughtDAG — Turn branching LLM conversations into a visible, editable thought graph. It's free. Best for conducting long, multi-branch reasoning sessions with LLMs.

ThoughtDAG is a free, open-source infinite canvas tool that transforms LLM conversations into an interactive, node-based thought graph where the connections between nodes represent the actual context passed to each subsequent request. Designed for researchers, developers, and power users who regularly engage in long, complex, or branching reasoning sessions with large language models, ThoughtDAG makes context visible, editable, and reproducible rather than buried in a linear scroll history. The tool connects to Ollama for fully offline, local-first use or to any OpenAI-compatible API endpoint, giving users flexibility over their model and privacy preferences. A hosted browser demo requires no installation and no signup, making it easy to evaluate immediately. Released under the MIT license with over 290 commits of active development, ThoughtDAG is a serious project for anyone who wants structured, transparent control over how AI reasoning unfolds across a session.

Key features

Infinite canvas for building and editing LLM conversation graphs
Visible, editable context wires connecting each conversation node
Supports Ollama for fully offline, local-first inference
Compatible with any OpenAI-compatible API endpoint

Pros & cons

PROS

+Completely free and MIT-licensed with no feature gating
+Runs fully offline via Ollama for privacy-sensitive workflows
+No install or signup required to try the hosted browser demo

CONS

−Graph-based interface has a steeper learning curve than a simple chat box
−GitHub adoption is limited so community support and ecosystem resources are still thin

Pricing

Free tier

All features included, fully free and open source

Paid from

Enterprise

Who is it for

→Conducting long, multi-branch reasoning sessions with LLMs
→Visualizing and auditing AI context flow across complex research tasks
→Building reproducible prompt chains for systematic experimentation

Frequently asked questions

Is ThoughtDAG free?

Yes, ThoughtDAG is completely free. It is released under the MIT open-source license and every feature is available at no cost, including offline use with a local Ollama instance.

What is ThoughtDAG best used for?

ThoughtDAG is best suited for long, branching LLM reasoning workflows where you need to see, edit, and control exactly what context is passed from one step to the next. It is ideal for researchers, developers, and analysts running multi-step or exploratory AI reasoning sessions.

What are the best alternatives to ThoughtDAG?

Comparable tools include FlowiseAI and LangFlow for node-based LLM pipeline building, Obsidian with AI plugins for graph-based knowledge work, and standard chat interfaces like Open WebUI for simpler Ollama interactions.

Is ThoughtDAG safe to use?

Yes. The source code is fully open and auditable on GitHub under the MIT license. When used with a local Ollama instance, all data stays on your own machine and nothing is sent to external servers.

How much does ThoughtDAG cost?

ThoughtDAG costs nothing. It is free and open source under the MIT license. Running it locally with Ollama incurs no API costs, and the hosted browser demo is also free with no signup required.

Reviews

No reviews yet. Be the first to review ThoughtDAG.