An independent AI architecture beyond Transformers and LLMs

GBR SYSTEM

Extreme Sudoku, 100% solved Without an LLM like ChatGPT, a sovereign AI that reasons from its own memory

GBR runs even on a palm-sized mini computer.

REASONING
Extreme Sudoku · Solve Rate 100%
MEMORY
4D fractal memory
ON-DEVICE DEMO
Jetson Orin Nano Super

REASONING RESULT

What LLMs could not solve, solved by Reasoning Only

From the 422,786 public test puzzles of Sudoku-Extreme, an ultra-hard Sudoku dataset, 10,000 were drawn by a rule fixed before any result was seen. GBR's reasoning engine solved every one in a single attempt, with no guessing or backtracking, and an answer certificate for every puzzle is on GitHub together with a checker anyone can run. (The certificates show that each answer is the unique solution of its puzzle, and we hold the run records showing that these answers came from the GBR reasoning engine. Because of concerns such as reverse engineering, the detailed data of the reasoning process can be viewed through an agreement such as an NDA.)

10,000 / 10,0000 wrong · all 10,000 puzzles drawn from the Sudoku-Extreme test set solved

7,379 puzzles never seen in development

All solved

726 standard puzzles from outside the dataset

All solved

An answer certificate for every puzzle

You can check it on GitHub

No guessing, no backtracking — each puzzle in one attempt
Home desktop PC specs · no GPU · 10,000 puzzles

ON-DEVICE DEMONSTRATION

On a small device, with no external server.

Filmed on an NVIDIA Jetson Orin Nano Super with 8 GB of memory, with the network cut off. It answers from stored knowledge and learns and retracts new knowledge on the spot. It moves a camera and a motor with natural language.

GBR SYSTEM / EDGEDEMONSTRATION
HARDWARE

NVIDIA Jetson Orin Nano Super

Execution demonstrated on the real device

Watch the demo on YouTube
The demo device — NVIDIA Jetson Orin Nano Super. Click the photo to open the video (14 min 53 s) in a new tab.
01
Offline questions and answers

Even with no network, it reads the knowledge stored on the device and answers with its sources.

02
New knowledge takes effect

Taught one sentence, it uses it from the very next question, with no restart or retraining.

03
Answers after retraction

Retracted knowledge is no longer used, and the record stays.

04
Camera and motor by natural-language command

It checks for safety before moving, then operates the servo motor and the camera.

PERSISTENT MEMORY

Knowledge, beyond the first answer.

Knowledge lives not in weights but in memory units with a source and a version. Add, correct or retract one and the very next answer reflects it, with no retraining, and what changed stays on record. When the evidence conflicts it asks back; when there is none it says it does not know.

It does not make up what it has not learned.

Demo video 8:08 — we asked about a company it had not learned.

GBR / MEMORY—
KNOWLEDGE / TAUGHT-01
ZIGU LABS — no memory

What is ZIGU LABS?

I have not learned that yet. I read it as "ZIGU LABS" — did I read it correctly?

The real conversation from the demo video (8:02–10:10), condensed from the on-screen text.

Watch this part of the video →

What the memory is like

4D fractal memory — a memory, a group of memories and a group of groups keep the same form, and a question follows that form down to read only the memories it needs.

4D fractal memory

The same form repeats at every size, from a single memory to groups and groups of groups, so even as memory grows, only the parts needed can be found and read. Our embedding is developed in-house too.

Search that misses nothing

Within a stated scope it finds every memory closest to the question without missing one; outside that scope it says so or holds the answer. This is Zero-Miss search.

Reads only what is needed

Knowledge is not held in RAM all at once: only the memories needed at that moment are streamed from storage, so a small device can use a large memory.

Updates without retraining

Publishing, correcting and retracting knowledge are all database operations. How long each memory is kept can be set individually.

01 / PERCEIVE
Perception

Images, sound and sensor input come in through our own perception engine.

02 / UNDERSTAND
Cognition

It understands the subject, the intent and meanings such as time, negation and conditions.

03 / REASON
Reasoning

It argues over memory by deduction and induction, and decides whether to answer, ask back or hold.

04 / GENERATE
Generation

It puts the decided meaning into words and links the evidence used.

05 / OUTPUT
Output

Out as speech, text and motion (actions later).

ARCHITECTURE

From perception to generation, on one mathematics.

Perception that sees and hears, cognition that understands, reasoning that argues by deduction, induction and more, and generation that composes speech all rest on the same mathematical framework of our own, and connect through the 4D fractal memory. It is a structure with no dependence on Transformers or LLMs.

Before speaking or acting, it first tests in an imagination space. With evidence it answers; when the evidence is split it asks back; without evidence it says it does not know.

GBR SYSTEM has two main parts. GBR DBMS, which stores and finds data GBR AI, which handles perception, cognition, reasoning, generation and output The two are connected around memory. All the engines are our own.

Through our own active foundation named INTEREST, metacognition and an autonomous learning loop, it aims to choose what to learn by itself and to verify what it has learned separately.

GBR SYSTEM architecture: under INTEREST, metacognition and the autonomous learning loop, perception → cognition → reasoning → generation → output, an imagination space above reasoning and generation, GBR DBMS persistent memory at the center (4D fractal memory, exact search, publish · correct · retract, streaming read, provenance receipts), and the mathematical foundation underneath
GBR SYSTEM architecture Open the diagram →

Compared with weight-based language models

Weight-based language modelGBR
ReasoningContinues text token by tokenReasoning over memory, tested in an imagination space before answering
Where knowledge is storedSpread across trained weightsVersioned memory units
State per stepA context window that grows with the conversation, up to a limitA fixed-size dialogue state, plus memory queried on demand
How learned results are applied and changedRetraining or fine-tuningPublishing, correcting and retracting memory units

APPLICATIONS / WHAT COMES NEXT

From your own AI to robots.

Even where there is no connection, the device sees, hears and decides on its own. Without an expensive server, a company or a person can put in the knowledge they need and have an AI of their own.

Small devices and robots

The first application area: one low-power small computer, no external server.

A platform for manufacturers

We are building a platform where manufacturers teach, change and use product knowledge on their devices. GBR PHYSICAL AI STUDIO, which connects device registration, licensing, updates, testing and reports, is its operating tool.

Cloud and device

Cloud and device run the same GBR. Learned memory moves to the device, and the two are designed to share memory with each other.

Sovereignty means control

An AI built and controlled by ourselves, not resting on foreign technology. On our own cognitive engine and memory store, companies and people manage their own knowledge and operating environment.

A CLOSER LOOK

Questions about GBR.

Is this a database attached to an LLM?

No. We built our own DBMS and our own AI and wove them together. Understanding questions and composing answers use no Transformer or LLM, and no external LLM is called.

How does it learn and correct knowledge? Does it need retraining?

All it needs is data. Through our own embedding and GBR AI, knowledge is stored as memory units with a source and a version. Publishing new knowledge, correcting it or retracting it are database operations, so the next answer reflects the change with no retraining. Our goal is to digest every kind of unstructured data.

What is the "4D fractal memory"?

A memory structure in which the same form repeats at every size, from a single memory to groups and groups of groups. Even as memory grows, only the part that is needed is found and read.

What is the "forgetting time"? Does GBR never forget?

By default, a memory lasts as long as it is not damaged. As storage fills up, however, memories come to need a time limit. You can set how long each memory is kept. It follows changes, and keeps what must be remembered for a long time.

Without weights, how does it give correct answers?

It finds what it needs in memory, then assembles and checks it in an imagination space. Correctness is kept in three layers: it finds the supporting memories exactly, verifies its reasoning with certificates, and tests in the imagination space before speaking. When evidence falls short, it says it does not know.

What problem does the "mathematics of our own" solve?

In short, the problems that arise in discrete updates and variational computation. It is a mathematical framework for the long-standing problems that arise when an AI updates and computes information. Perception, cognition, reasoning, memory and generation all stand on it.

How can I check the Sudoku results?

Download the original dataset (from Hugging Face) and run the checker from the GitHub repository, and anyone can use the answer certificates to re-check that each of the 10,000 answers is the unique solution of its puzzle. All 10,000 were also verified with an independent checker we did not write. The certificates show that the answers are correct; that these answers came from the GBR reasoning engine rests on our internal run records.

What does it run on? Does it need the internet?

The current demo ran on a single NVIDIA Jetson Orin Nano Super with 8 GB of memory, with the network cut off. Knowledge lives in memory on the device, so it answers with no external server and handles devices such as a camera and motor in natural language. Smaller and larger devices are planned next.

What can GBR do today, and what are its limits?

It still has much to learn. It answers questions from knowledge learned in numbers and Korean, and learns and retracts new knowledge through conversation. It has not learned much data yet, so it is not as fluent as LLMs or as striking at generation, but it clearly works. (It is built in Korea, so for now it has learned only Korean and numbers.)

Who do you work with?

Makers of small devices and robots, and researchers in machine learning, databases, speech and robotics. Together we define the task, the device constraints, the memory and interfaces needed, and the scope of a technical review and a PoC.

BUILD WITH ZIGU LABS

Join an infinite journey that begins on Earth. We are still waiting for you.

From technical questions and research to product partnerships — let's talk about knowledge and intelligence that need sovereignty.

agent@zigulabs.comGBR SYSTEM / ZIGU LABS