MLCommons Association
MLCommons is an open engineering consortium with a mission to accelerate AI innovation and increase its positive impact on society through industry-standard benchmarks and datasets.
At a Glance
- Semiconductor manufacturers
- Cloud service providers
- AI software developers
- Academic research institutions
- +1 more
AI Tools by MLCommons Association
(1)
MLCommons
Open Source AI Benchmark Suite
Discussions
No discussions yet
Be the first to start a discussion about MLCommons Association
Latest News
MLCommons Releases MLPerf Client v1.6 with Performance Optimizations and Enhanced User Experience
A new GPT-OSS benchmark and DeepSeek R1 updates for latency-optimized reasoning
Standardizing Generative AI Service Evaluation: An API-Centric Benchmarking Approach
YOLO for the MLPerf Inference v6.0 Edge Suite
Products & Services
Benchmark suite for measuring ML training speed to a target quality metric.
Benchmarks for measuring inference speed in datacenter, edge, mobile, and tiny devices.
One of the largest open-source speech-to-text datasets with over 80,000 hours of data.
An infrastructure-agnostic framework for shipping machine learning models.
Market Position
MLCommons is the globally recognized authority for vendor-neutral AI hardware and software benchmarking, providing the 'Gold Standard' for industry performance metrics.
Leadership
Founders
Peter Mattson
Senior Staff Engineer at Google; previously founded the Programming Systems and Applications Group at Nvidia Research; PhD/MS from Stanford.
David Kanter
Semiconductor and computing expert; founder of microprocessor and compiler startups; Head of MLPerf; BS/BA from University of Chicago.
Debojyoti Dutta
AI Solutions and Engineering Lead at Nutanix; former Distinguished Engineer at Cisco; PhD in CS from USC.
Vijay Janapa Reddi
Associate Professor at Harvard University; researcher in computer architecture for autonomous and mobile computing; PhD from Harvard.
Carole-Jean Wu
Research Scientist at Meta (Facebook AI Research); specialist in energy-efficient systems and ML execution at scale; PhD from Princeton.
Executive Team
Rebecca Weiss
Executive Director
Former Head of Research & Innovation at Mozilla; AI policy advisor to U.S. Congress; PhD from Stanford.
Peter Mattson
President
Senior Staff Engineer at Google; founder of MLPerf; PhD from Stanford.
Board of Directors
Founding Story
Initially formed as the MLPerf consortium in 2018 by engineers from Baidu, Google, and top academic institutions to create standardized AI performance metrics. It formally launched as the non-profit MLCommons Association in December 2020 to expand its scope into datasets and infrastructure.
Business Model
Revenue Model
Membership dues from corporate, academic, and non-profit members.
Pricing Tiers
For-profit organizations with 500+ full-time employees. Includes $90,000 initiation fee.
For-profit organizations with 10-499 full-time employees. Includes $18,000 initiation fee.
Faculty at non-profit academic institutions.
Individual participants with board approval.
Target Markets
- Semiconductor manufacturers
- Cloud service providers
- AI software developers
- Academic research institutions
- Government agencies
- AI hardware performance comparison
- AI model training efficiency
- Safety and reliability testing
- Clinical AI evaluation
- NVIDIA
- Intel
- AMD