karpathy

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

sudoingX

Qwen 3.5 9B -- Thermodynamic Computing Reasoning Test

9 billion parameters reasoning about bleeding-edge physics on a $200 GPU. Full thinking chain visible. Nothing cherry-picked.

Hardware

Field	Value
GPU	RTX 3060 12GB
Model	Qwen 3.5 9B Q4_K_M (5.3GB)

karpathy

"""
The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
This file is the complete algorithm.
Everything else is just efficiency.

@karpathy
"""

import os       # os.path.exists
import math     # math.log, math.exp

uubzu

// For use on https://twitter.com/settings/your_twitter_data/twitter_interests
function uncheckAll() {
    let inputs = document.querySelectorAll('input[type="checkbox"]');
    let delay = 0 
    for (let i = 0; i < inputs.length; i++) {
        if (inputs[i].checked) {
            setTimeout(() => { inputs[i].click( new Event('change') ) ; console.log(i) }, delay);
            delay += 1000;
        } else {
            console.log("Skipped " + i)

jart

#!/bin/sh
# rename-pictures.sh
# Author: Justine Tunney <jtunney@gmail.com>
# License: Apache 2.0
#
# This shell script can be used to ensure all the images in a folder
# have good descriptive filenames that are written in English. It's
# based on the Mistral 7b and LLaVA v1.5 models.
#
# For example, the following command:

cpursley

PostgreSQL is Enough

See the new site: https://postgresisenough.dev

---

• Simplify: move code into database functions
• Just Use Postgres for Everything
• Just use Postgres
• PostgreSQL is the worlds’ best database

adrienbrault

# Clone llama.cpp
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp

# Build it
make clean
LLAMA_METAL=1 make

# Download model
export MODEL=llama-2-13b-chat.ggmlv3.q4_0.bin

aallan

import network
import socket
import time
import struct

from machine import Pin

NTP_DELTA = 2208988800
host = "pool.ntp.org"

sandyjmacdonald

SSID = ""
PSK = ""

ritchieking

sqlcsv() {
  if [ $# -lt 2 ]
  then
    echo "USAGE: sqlcsv [filename.csv] [SQL]"
    echo "In the SQL, refer to the data sourse as [filename]"
  else
    filename="$1"
    dataname=${filename%????}
    sqlite3 :memory: -cmd '.mode csv' -cmd ".import $filename $dataname" -cmd '.mode markdown' $2
  fi