Graph-Oriented Generation – Beating RAG for Codebases by 89%
LLMs are better at being the "mouth" than the "brain" and I can prove it mathematically. I built a deterministic graph engine that offloads reasoning from the LLM. It reduces token usage by 89% and makes a tiny 0.8B model trace enterprise execution paths flawlessly. Here is the white paper and the reproducible benchmark.
In plain words
Graph-Oriented Generation is a system that uses a deterministic graph engine to handle reasoning tasks that are normally performed by large language models. Instead of relying on the LLM for complex logic, it offloads this work to the graph component, allowing smaller models to process enterprise execution paths accurately. The approach reduces token usage significantly and is designed for developers working with codebases who need efficient code understanding and tracing without the computational overhead of traditional retrieval-augmented generation methods.
written from the facts on this page · September 2026
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
