byte-level

A tiny byte-level multi-head content classifier (~1.5M params, ~200KB ONNX, <6ms). Classifies code, text, markup, config, images, binary, secrets, 62 code languages, 30 text languages, 90 MIME types from raw bytes — no tokenizer needed.

An editable, auditable 807K-param byte-level LLM: CRUD single facts with provable per-edit locality, and abstain when unsure instead of guessing. CPU, offline.

Kronecker Embeddings: byte-level structured token representations for parameter-efficient language models. Reference implementation.

🐍This is a fast, lightweight, and clean CPython extension for the Byte Pair Encoding (BPE) algorithm, which is commonly used in LLM tokenization and NLP tasks.

An efficient openblocks parser module.

A deterministic byte-level BPE tokenizer in pure Python, built from scratch with strict tests, typed code, and polished docs.

Tiny byte-level AI on CPU: zero-dependency C++20 inference engine and neural decision models (2.8ms latency, 9.4MB RAM). Replaces tokenizers with a 256-byte vocabulary. Flagship specialists for pre-commit secret scanning (0.797 F1 vs 0.337 GitLeaks) and PII redaction. Apache 2.0.

File security system using remote authentication

Tokenizer-free HSL text-to-speech feasibility prototype with open code and weights. Working synthesis evidence; formal perceptual evaluation remains open.

Four-seed matched ASR feasibility study using an HSL byte-signal substrate and spectral lens. Open code, weights, controls, and limitations; not SOTA.

The official training stack for PureByte: train byte-level neural decision specialists from a task definition to a verified GGUF model in ~25 minutes on a single GPU. Includes synthetic hard-negative generation, curriculum training, and automated evaluation against CredData and PIIMB.

PyTorch transformer input from raw bytes with zero learned input parameters: no tokenizer, embedding table, or learned input projection.

Engram without a tokenizer

Open HSL/HoLo research: a deterministic byte-native signal substrate tested across language, speech, and diffusion with matched controls and honest limits.

NeuralPiece: An adaptive byte-level neural tokenizer designed to surpass traditional BPE and Unigram via deep learning chunking.

Dragon Hatchling Architecture based Neural Language Model

Language-model proof of operation for HSL's zero-learned-parameter input path. Open code, weights, and documented limitations; not a SOTA claim.

Deterministic, non-learned byte-to-HSL signal encoder for PyTorch: 27-D interpretable features, lossless byte codec, tests, controls, and GPU paths.