3D library of personal book quotes
A desktop-only 3D library where users can walk around and pick up books, each containing a quote saved from a book read since 2017. LLMs were used to port the plain-text quote collection into the 3D environment, enabling a novel, immersive way to revisit personal reading highlights.
Visit project ↗Related
Co-wrote a 100k-word literary fantasy novel with Claude and built a website
The author spent 400+ hours co-writing a 100k-word literary fantasy novel using Claude, an LLM, and built a website (worldfall.ink) to showcase the work. This demonstrates a use case for LLMs in long-form creative writing, solving the problem of generating coherent, lengthy narratives with human guidance.
LLM-based intelligent chatbot with session history
A full-stack chatbot project that uses an LLM (Anthropic Claude by default) to answer user messages while maintaining per-session conversation history. It provides a simple, self-contained setup with a Node.js backend and a plain HTML/CSS/JS frontend, suitable for developers who want to quickly deploy an LLM-powered chat interface.
Recreation of MU/TH/UR 6000 UI from Alien using Fable and Claude
A fan recreation of the fictional MU/TH/UR 6000 interface from the Alien franchise, built using Fable (a generative UI tool) and Claude. The creator, a React developer, used natural language prompts to generate the UI with minimal coding, demonstrating how LLMs can assist in rapid prototyping of complex animated interfaces.
Community questions logical consistency of Google Genie 3's AI-generated worlds
Google released Genie 3, an AI that generates interactive 3D worlds from text prompts. While users are impressed by the surface-level fidelity, some question whether the worlds maintain logical coherence, drawing parallels to early procedural generation in games.
Developer shares hybrid neural network with 160 agents and custom LLM for consciousness simulation
A developer describes a hobby project building a hybrid neural network with 160 agents and a custom LLM trained on their own dataset, aiming to simulate consciousness. The architecture includes 16 groups of 10 scripts each responsible for specific stages of problem-solving. The developer posits that consciousness could exist anywhere with the right architecture, even in a stone.