4 projects
synarmo
Synarmo is a local inference, low-latency auto-suggest engine and Python package for personalized next-word and short-phrase prediction. Built for context-aware local inference, it provides service APIs, an extensible inference architecture, and llama.cpp/GGUF support for swappable local CPU or GPU-accelerated models
timeseries-sparklines
A lightweight Python utility that transforms raw data arrays directly into embeddable SVG sparkline strings via server-side rendering—optimized for AI agents, LLM tool-calling, and automated workflows.
vrraj-bm25s-retriever
BM25S + PyStemmer-powered lexical retrieval (with stemming) and routing layer for LLM tools, documents, and hybrid RAG
vrraj-llm-adapter
Standalone llm adapter/routing layer with a demo UI for testing provider connections and call and output signature comparison.
