Skip to content
@purebyte-ai

Purebyte

Not all intelligence needs to be massive. Building the foundation for private, local, byte-level AI specialists.

Pinned Loading

  1. purebyte purebyte Public

    Tiny byte-level AI on CPU: a zero-dependency C++17 runtime and neural decision models (0.8 ms per 64-byte decision, 9.4 MB RAM). Reads raw bytes, no tokenizer. Specialists for secret scanning (F1 0…

    C++ 3 1

  2. purebyte-train purebyte-train Public

    The training stack for PureByte: from a task definition to a verified GGUF specialist in 25 minutes to an hour on one consumer GPU. Synthetic look-alikes grounded in real bytes, leak checks against…

    Python

Repositories

Showing 2 of 2 repositories
  • purebyte-train Public

    The training stack for PureByte: from a task definition to a verified GGUF specialist in 25 minutes to an hour on one consumer GPU. Synthetic look-alikes grounded in real bytes, leak checks against the exams, criteria frozen before training, multi-seed runs with controls, and proof that the C++ runtime gives the same answers.

    purebyte-ai/purebyte-train's past year of commit activity
    Python 0 Apache-2.0 0 0 0 Updated Sep 28, 2026
  • purebyte Public

    Tiny byte-level AI on CPU: a zero-dependency C++17 runtime and neural decision models (0.8 ms per 64-byte decision, 9.4 MB RAM). Reads raw bytes, no tokenizer. Specialists for secret scanning (F1 0.797 vs 0.337 for gitleaks on CredData) and PII redaction. Code under Apache-2.0; model weights under their own license.

    purebyte-ai/purebyte's past year of commit activity
    C++ 3 Apache-2.0 1 0 0 Updated Sep 28, 2026

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Loading…

Most used topics

Loading…