8 projects
winnex-madhava
Deterministic vector search with Cauchy-Schwarz bound pruning + exact-L2 post-filter (BIGANN L2 verified) — provable completeness for RAG indexing and regulated retrieval.
winnex-ai-normalize
Input normalization plug for the Madhava engine — embed text via any provider (OpenAI, Qwen3, nano, xfactor), validate/normalize to float32/uint8, with provider failover and NO fake-vector fallback. Consumed by winnex-madhava, winnex-tracer, Liferay and the Maestro.
winnex-xfactor
Deterministic manifold embedding for native LLM inference — the X-Factor operator (O(D²r) power iteration, no training).
winnex-nano
Native Winnex inference engine — deterministic spectral tokenizer, multimodel weight balancing, and chunk streaming via the Madhava fast build. Model-agnostic (Qwen/BERT/DeepSeek/GPT), OpenCL, no CUDA.
winnex-tracer
The unified Winnex tracer family — shared core (WORM, Ed25519 signing, AuditCommitment) plus the gov/med/jus/gap domains, in one namespaced package.
winnex-tracer-core
The shared base for the Winnex tracer family — WORM storage, Ed25519/ICP signing, tamper-evident audit log, retention policies and the AuditCommitment (non-repudiation). Reused by tracer-gov, tracer-med, tracer-jus and tracer-gap.
winnex-madhava-sec
Mathematically Guaranteed Agent Security Framework — Cauchy-Schwarz bound pruning for AI agent prompt-injection detection, attack amplification, and multi-embedder semantic safety. PiPrime navigation + bounds + SafetyEnsemble.
madhava-l2
Deterministic vector search with Cauchy-Schwarz bound pruning + exact-L2 post-filter (BIGANN L2 verified)
