MLCube® is a project that reduces friction for machine learning by ensuring that models are easily portable and reproducible.

Quantitative benchmark for llm safety filter effectiveness using MLCommons AI Safety taxonomy

A Go implementation for working with the ML Commons Croissant metadata format.

This repository contains the spreadsheet of the quantitative analysis performed for the paper "Suitability of Forward-Forward and PEPITA Learning to MLCommons-Tiny benchmarks".

mlcommons tiny performance benchmark on EBAZ4205 (former bitcoin mining board built around Xilinx Zynq-7000 SoC)

As described in "Towards Full On-Tiny-Device Learning: Guided Search for a Randomly Initialized Neural Network"

LLM Billing & Benchmarking Standard (LBBS) v0.1 — Draft for public comment.

Transparent benchmark cards, JSON artifacts, and visual tools for photonic AI accelerator energy/noise claims.

Multi-turn safety testing plugin for MLCommons ModelBench. By Anivar Aravind.

Reference implementations of MLPerf® inference benchmarks

🌱 Community-driven dataset projects for inclusive, safe, and open AI engineering.

💰 Establish a standard for LLM billing and benchmarking to enable fair comparison of models with clear metrics and compliance levels.