suncityldp / zx-bench

A reproducible local LLM benchmarking platform with versioned datasets, deterministic scoring, and auditable reports.

mrtinkz / ai-charter

Universal AI Charter: plain-language governance principles and a public certification registry for AI models, built on disclosure instead of a development slowdown.

microsoft / scope

Scope | Agentic Experience evaluation platform (Research preview)