BDH (Dragon Hatchling) is a biologically inspired large language model architecture that connects principles of deep learning with the foundations of neuroscience. Developed by researchers at Pathway, BDH provides a theoretical and practical framework for understanding the emergence of reasoning and generalization in artificial systems.
This repository contains the official implementation from the paper:
A. Kosowski, P. Uznański, J. Chorowski, Z. Stamirowska, M. Bartoszkiewicz.[, arXiv (2025).]The Dragon Hatchling: The Missing Link between the Transformer and Models of the Brain
BDH represents a scale-free, locally interacting network of neurons capable of intrinsic reasoning dynamics. BDH scales like a Transformer on performance benchmarks—yet retains full interpretability and theoretical grounding in the fine-grained dynamics of neuron interactions.
Key properties:
Scale-free network topology mimicking biological connectivityLocally interacting neuron particles with excitatory/inhibitory dynamicsHebbian working memory based on synaptic plasticity, displaying monosemanticityGPU-friendly state-space formulation for efficient implementationInterpretable activations that are sparse and positive
BDH formalizes a bridge between neural computation and machine-based language understanding. It shows how macro reasoning behavior in large AI models emerges from micro-level neuron dynamics, guided by principles of graph theory and local computation.
Empirically, BDH matches GPT-2–scale Transformers across language and translation tasks at equivalent parameter scales (10M–1B).
BDH and the Transformer share attention-inspired computation; however, BDH’s graph-based architecture makes its attention emerge naturally from neuron-level interactions, reflecting attention as seen in biological systems.
BDH follows Transformer-like scaling laws, maintaining parameter efficiency while achieving interpretability at any scale.
Note: The Sudoku Extreme result refers to Pathway’s internal BDH implementation, not to the current open-source repository. This repository contains the implementation of the baseline variant as described in our public paper and does not reproduce the 97.4% benchmark result out of the box. See the dedicated Extreme Sudoku research blog post for additional benchmark context and the reported results.
On Sudoku Extreme, BDH reaches 97.4% accuracy across roughly 250,000 difficult puzzles, without chain-of-thought, solution backtracking, or external tool use, while leading LLMs struggle to perform on the benchmark at all.
Language is not enough for intelligence. Transformers process information token by token with limited internal state, which makes search-heavy, non-linguistic reasoning tasks like Sudoku awkward. BDH uses a larger latent reasoning space with intrinsic memory that supports learning and adaptation during use.
We believe that the future of AI will belong to systems that can reason natively across domains, that can hold multiple possibilities in a rich latent space, and that can converge on solutions without needing to verbalize every step. BDH is our answer to that challenge. It is designed to be a universal reasoning system that can speak our language without being trapped inside it. And yes, it solves Sudoku.
Read more: Post-transformers: Sudoku Bench
| Model | Sudoku Extreme Accuracy | Relative Cost |
|---|---|---|
| Pathway BDH | 97.4% | 10× lower, No chain-of-thought |
| Leading LLMs (O3-mini, DeepSeek R1, Claude 3.7 8K) | ~0% | High (chain-of-thought) |
Table 1: Performance comparison on extreme Sudoku benchmarks (~250,000 difficult puzzles).
Source: Pathway internal data and https://arxiv.org/pdf/2506.21734 for the Leading LLMs’ accuracy score. Pathway’s approach reflects top-1 accuracy and does not rely on chain-of-thought nor solution backtracking.
pip install -r requirements.txt
python train.py
Watch the
SuperDataScience podcast(72 min.) featuring Adrian Kosowski in conversation with Jon Krohn, unpacking BDH’s neuron-level architecture and sparse reasoning dynamics.▶️ Dragon Hatchling: The Missing Link Between Transformers and the Brain - Read about BDH in
,Forbes,Semafor,The Turing Post,Quantum Zeitgeist, and elsewhere in the media.Golem - Discuss and share the BDH paper on:
,Hugging Face Papers, andAlphaxiv.EmergentMind
adamskrodzki/bdh: dynamic vocabulary, stateful attentionmosure/burn_dragon_hatchling: Burn portseverian42/bdh: MLX portGit-Faisal/bdhGrahLnn/bdh
We thank Andrej Karpathy for the nanoGPT code and the tiny Shapespeare dataset used in this demonstration.
BDH research stands at the intersection of AI architecture, biological learning models, and theoretical computer science—an effort to map the equations of reasoning between artificial and biological intelligence.