Apache 2 Python Packages

attestplane

Verifiable audit substrate for AI agents — EU AI Act Article 12 ready. Apache 2.0.

diarize

Speaker diarization for Python — "who spoke when?" CPU-only, no API keys, Apache 2.0. ~10.8% DER on VoxConverse, 8x faster than real-time.

s4-codec

GPU-accelerated transparent compression S3-compatible storage gateway. Drop-in replacement for AWS S3 endpoints; cuts your S3 bill 50-80% with no app changes (Rust, nvCOMP, zstd).

s4fs

GPU-accelerated transparent compression S3-compatible storage gateway. Drop-in replacement for AWS S3 endpoints; cuts your S3 bill 50-80% with no app changes (Rust, nvCOMP, zstd).

knowlytix-core

Geometric Memory Systems — core geometry and LLM-layer foundations

predacore

PredaCore — your personal autonomous agent with an infinite context window and memory.

sentinel-kernel

Evidence infrastructure for the regulated AI era. Trace every decision, attest it cryptographically, audit the record, comply with EU AI Act Art. 12/13/14/17 via a signed evidence pack. Air-gapped, offline-first, Apache 2.0.

signet-sign

Capability-based safety gates for LLM agents. The model proposes; signet authorizes.

alphadynamics

Compact sequence-only neural propagator for protein torsion dynamics. 2.39x lower JSD than Microsoft Timewarp at 3000x fewer parameters. pip install alphadynamics.

knowlytix-benchmark

Benchmark for structured information retrieval from financial documents using graph-verifiable questions

matyan-api-models

Open-source ML experiment tracking system built on FoundationDB — a self-hosted alternative to Aim/AimHub with a Kafka-based ingestion pipeline and cloud artifact storage.

knowlytix-knowledge

GMS-powered document ingestion, query, and learning — DocGMS

singleaxis-fabric

Open-source observability and control plane for AI agents — decision-span tracing, fail-loud guardrails, and human-in-the-loop, OpenTelemetry-native.

knowlytix-harness

GMS-Harness — provider-agnostic DOE-driven black-box testing platform for LLM agents

tessera-llm-proxy

Drop-in LLM cost-optimization proxy. Auto-route + cache + compress + batch. Flat monthly pricing by token volume, keep 100% of savings. Free 60M tokens/mo.

german-ocr

German-OCR is specifically trained to extract text from German documents including invoices, receipts, forms, and other business documents.

wiseorder

WiseOrder — the kernel that governs how an AI's output is allowed to become an action, and seals every decision. Four-class verification, hash-chained audit memory, conformance vectors, independent Go + Rust verifiers. A Wise.Est Systems protocol.

matyan-ui

Matyan web UI server — serves the React app and static assets

matyan-backend

Production-ready experiment tracker

matyan-frontier

Ingestion gateway: WebSocket + presigned blob URLs, publishes to Kafka

llmesh-lleval

Reserved name — lleval: LLM eval framework for the FullSense LLM family (on-prem + cloud unified A/B, progressive size matrix, honest disclosure). Reference implementation in development.

urml-llm-bridge

The specification, reference runtimes, validator, LLM bridge, and conformance suite for URML — an open language for robot intent.

matyan-client

Matyan SDK — Aim-compatible client for experiment tracking

agentboot-ai

AI agent for bare-metal hardware detection and OS installation — conversationally