LOCAL-FIRST · 2,489 conversations · 0 cloud calls

I compiled my own ChatGPT
into a behavior graph.

The Sovereign Knowledge Compiler reads every conversation export, runs a local LLM (deepreinforce-ai_Ornith-1.0-35B-Q4_K_M.gguf) to distill themes and surface the why, and stores the result as a static, inspectable artifact. This page is the live proof — built entirely from what it generated. No data leaves your machine at runtime.

compiled in 2163.8s · model deepreinforce-ai_Ornith-1.0-35B-Q4_K_M.gguf · 2,489 conversations → 120 themes · 2023-03-12 → 2026-07-16
2,489
conversations compiled
19,435
message turns
45
multimodal chats
441
forked threads
120
explored themes
59
images extracted

Conversation Graph · 3D

drag to rotate · scroll to zoom · hover a node
nodes = conversations + themes · size = turns / frequency · edge = belongs-to / co-occurs · color = theme

Themes explored

prompt-engineering109
creative-writing71
mental-health54
ollama53
content-creation46
geopolitics41
image-generation39
grief38
nextjs34
web-development34
python32
debugging27
stable-diffusion27
machine-learning25

Activity over time

■ conversations■ turnsper month

Deepest conversations

Ranked by derived importance: turn depth + thread forks + multimodal − archived. The "what I kept coming back to" signal.

SaaS Development Environment Setup
—
saas-developmentdjango-setupdatabase-configuration2024-10-11⑂ 16 forkso1-mini

A note on sovereignty

This graph was compiled from a raw ChatGPT export that left OpenAI's servers and became your artifact. The reasoning happened once, at compile time; this page only reads static files. Nothing is sent anywhere when you browse it.