Fractal Media Infrastructure
Selected active lanes
This page highlights current active lanes rather than attempting to reproduce the full repository index. For the complete GitHub corpus, use the instance001 repository surfaces directly.
Public-facing surfaces for understanding FMI, finding support, checking app-store releases, and choosing a path into the broader repository ecosystem.
Examples: FMI Pages Site, Public Record, Whatisthisgithub, Easy Start Guide, FMI Research Papers, Google Play releases
Projects focused on offline-capable assistants, small-phone Android chat, local media generation, local LLM build and training workflows, and practical software for everyday use. The stance is local first, cloud optional: cloud when it helps, local when it fits.
Examples: Chatty-Cog, Chatty-mini, Chatty-Pet, Chatty-Art, Chatty-Art Windows install, Chatty-Lora, Nanochat LLM Tweaker, Chatty-Factory, ChattyFactory Rebuild, Chatty-Quest
Projects focused on accessible, local-first educational software for students, teachers, parents, and self-directed learners who need practical tools without accounts or tracking. Local-first here means user control first, with hosted services kept as deliberate choices rather than hidden dependencies.
Examples: Chatty-EDU, Chatty-EDU Windows install, Chatty-EDU User
Reusable guidance for higher-quality human-AI collaboration, session recovery, drift reduction, and explicit operating constraints across local models, hosted models, tools, and providers.
Examples: AI Teaming Framework, Model Behaviour Toolkit
Research into host-side memory, relational reasoning, earned abstraction, failure-aware routing, reducer-governed persistence, semantic sorting, behavioural suitability profiling, and compact continuity for legible local AI systems. Project Leviathan also develops an exploratory cognitive-cartography method: controlled withheld-conclusion tests and cross-domain mapping intended to measure reachable reasoning structure rather than assign a single reasoning score.
Examples: Project Leviathan, EF Engine, RD Engine, LLM Semantic Dataset Sorter, Cognition Mesh Test Chamber, LLM-Defined Persistent Memory, Semantic Signal Alphabet
Work examining alignment incentives, epistemic governance, safety theater, and the practical consequences of suppressing user or operator agency.
Examples: Governance by Design Report, Safety Theatre
A public policy companion for model choice, local/cloud hybrid control, proportionate restraint, broad national capability, and grassroots participation in Australian AI sovereignty.
Examples: AI Fair-Go page, Australian AI Fair-Go repo, Chatty-EDU, Chatty-Cog
Long-cycle conceptual work on cognition, entropy-handling, model behavior, and associated theoretical scaffolds intended for public scrutiny rather than institutional gatekeeping.
Examples: Relational Curriculum Geometry, Entropy Folding hypothesis, cross-domain signal atlas, Project Leviathan, foundational frameworks