Core is a self-contained reasoning substrate. Its state changes as it is used. Each query enters a persistent field that already holds concepts, constraints, bindings, examples, procedures, corrections, and unresolved candidates.

Active structures in that field compete for support. Well-supported ones strengthen, conflicts get resolved or carried, and weakly supported candidates stay provisional. Core returns the answer together with the formation that produced it: what activated, what conflicted, what resolved, what remains missing, and what changed for the next query.

Core can be reached through chat, API calls, agents, logs, or embedded workflows. These surfaces format and relay information. The reasoning happens in the substrate underneath them.

We build Core, an adaptive reasoning substrate that learns from use, preserves useful task structure, and revises it as evidence changes.

We build Core for systems that need to learn while they operate.

State carries forward: Later work can use constraints, relationships, and evidence formed earlier. Returned outcomes can revise later behavior: Admitted observations and measured results can change which task structure carries support. Results stay inspectable: Your application receives defined outputs, supporting evidence, and learned state.

$REI is the ecosystem token of REI and Unit 00. It has been part of the project from the beginning and will remain central to its economic layer.

Enterprise customers can run Core instances over valuable data and let them accumulate useful knowledge and task structure through use. We plan marketplaces where owners can offer access to those instances or publish selected knowledge, learned structures, and inspectable traces for others to use.

Only material that an owner chooses to make available will enter these markets. Access to instances and exchange of published artifacts are planned to settle in $REI. This gives instance owners a way to capture value from useful machine work while retaining control over what they share.