About Us

Meta's AI & Compute Foundation team, led by CQ Tang (a.k.a. Chunqiang Tang), comprises world-class research scientists and engineers developing foundational AI and compute technologies. Our focus areas include:

• Drive Meta's strategy for hardware innovation: We define and execute the company's strategy for new hardware technologies, spanning Meta's homegrown silicon---AI chips, custom ARM CPUs, and CXL ASIC---as well as vendor-supplied GPUs, CPUs, memory, and storage.
• Productionize innovation: We translate innovation into real-world impact by co-designing across every layer of the stack---from silicon to software and AI models---creating integrated solutions that redefine performance at scale.
• Build AI software to serve billions: We develop the complete software ecosystem for Meta's homegrown AI chips, ranging from compilers and kernels to large-scale LLM inference and training. We also build foundational software technologies for GPUs, CPUs, and hyperscale infrastructure.

Our primary measure of success is delivering AI and compute systems into production to serve billions of people across Facebook, Instagram, WhatsApp, Messenger, Threads, and Meta AI. This requires continuous innovation, and our philosophy is "research in production": first delivering hyperscale production systems with real-world impact, then summarizing our innovations and experiences as research papers at top conferences.

Papers based on our hyperscale production systems have won prestigious Best Paper Awards at SOSP, OSDI, ISCA, and ASPLOS, and have been featured as IEEE Micro Top Picks, Communications of the ACM (CACM) Research Highlights, and a CACM cover story. We also publish in broader venues such as ICML, NeurIPS, SC, MICRO, HPCA, NSDI, VLDB, and MLSys.

While we value our strong publication record, we take even greater pride in the global impact of the production systems we build for billions of people.

A Peek Into Our Work

The broad scope of our work---spanning hardware, software, and AI models---makes a complete overview challenging. A quick way to glimpse our work is by reading the abstracts of the few papers highlighted below. For broader topics, please refer to our publications.

• [Communications of the ACM] Meta's Hyperscale Infrastructure: Overview and Insights
• [Meta AI Blog] Four MTIA Chips in Two Years: Scaling AI Experiences for Billions
• [OSDI'26] Hardware Lifecycle-Aware Power Planning in Commercial Hyperscale Datacenters
• [ISCA'26] MTIA-300: Meta's Training Chip with Embedded NIC Chiplets and Communication Offloading Engine
• [ISCA'25] Scaling Llama 3 Training with Efficient Parallelism Strategies
• [ISCA'25] DCPerf: An Open-Source, Battle-Tested Performance Benchmark Suite for Datacenter Workloads
• [ICML'24] Wukong: Towards a Scaling Law for Large-Scale Recommendation
• [ISCA'23] Shared Microexponents: A Little Shifting Goes a Long Way
• [ISCA'23] TPP: Transparent Page Placement for CXL-Enabled Tiered-Memory

2026

• [OSDI'26] Hardware Lifecycle-Aware Power Planning in Commercial Hyperscale Datacenters
• [OSDI'26] Syncopate: Automatic Fine-Grained Compute-Communication Overlap via Chunk-Centric Scheduling
• [ISCA'26] MTIA-300: Meta's Training Chip with Embedded NIC Chiplets and Communication Offloading Engine
• [ISCA'26] Power Sloshing in Compound Servers for Large-Scale AI Inference Workloads
• [ISCA'26] Vistara: Making CXL Real--Full Path from ASIC Design and OS Support to Hyperscale Deployment
• [ISCA'26] LoKA: Low-precision Kernel Applications for Recommendation Models At Scale
• [ASPLOS'26] Triton-Sanitizer: A Fast and Device-Agnostic Memory Sanitizer for Triton with Rich Diagnostic Context
• [ASPLOS'26] SNIP: An Adaptive Mixed Precision Framework for Subbyte Large Language Model Training
• [ASPLOS'26] SEVI: Silent Data Corruption of Vector Instructions in Hyper-Scale Datacenters
• [HPCA'26] ReScue: Reliable and Secure CXL Memory
• [HPCA'26] PinDrop: Breaking the Silence on SDCs in a Large-Scale Fleet (Best paper finalist)
• [HPCA'26] RoMe: Row Granularity Access Memory System for Large Language Models
• [MLSys'26] Optimizing Deployment Configurations for LLM Inference
• [MLSys'26] Demystifying the Mixture of Experts Serving Tax
• [MLSys'26] Sparing Strategy to Minimize Reliability Impact on Large Scale Training Jobs
• [MLSys'26] Agentic Operator Generation for ML ASICs
• [S&P'26] SoK: Systematizing a Decade of Architectural RowHammer Defenses Through the Lens of Streaming Algorithms
• [IEEE Micro Journal] Demystifying the cost versus benefits of sparse LLM acceleration
• [IEEE Micro Journal] Triton for MTIA: Bridging the Programming Model Gaps for Custom AI Accelerators
• [Transaction on Computer Systems (TOCS)] Detecting Tiny Performance Regressions at Hyperscale
• [Communications of the ACM (CACM) Research Highlights] TMO: Transparent Memory Offloading in Datacenters

2025

• [OSDI'25] WLB-LLM: Workload-Balanced 4D Parallelism for Large Language Model Training
• [SOSP'25] Mercury: Unlocking Multi-GPU Operator Optimization for LLMs via Remote Memory Scheduling
• [ISCA'25] Meta's Second Generation AI Chip: Model-Chip Co-Design and Productionization Experiences
• [ISCA'25] DCPerf: An Open-Source, Battle-Tested Performance Benchmark Suite for Datacenter Workloads
• [ISCA'25] Scaling Llama 3 Training with Efficient Parallelism Strategies
• [MICRO'25] Re-architecting End-host Networking with CXL: Coherence, Memory, and Offloading
• [MICRO'25] Multi-Dimensional ML-Pipeline Optimization in Cost-Effective Disaggregated Datacenter
• [MICRO'25] Chameleon: Adaptive Caching and Scheduling for Many-Adapter LLM Inference Environments
• [HPCA'25] Machine Learning-Guided Memory Optimization for DLRM Inference on Tiered Memory
• [Communications of the ACM] Meta's Hyperscale Infrastructure: Overview and Insights
• [ARITH'25] An Empirical Study of Microscaling Formats for Low-Precision LLM Training
• [MLSys'25] Context Parallelism for Scalable Million-Token Inference
• [MLSys'25] Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Training

2024

• The Llama 3 Herd of Models
• [IEEE Micro'24] End-to-End Cloud Application Cloning With Ditto Top Picks
• [IEEE Micro'24] Contiguitas: The Pursuit of Physical Memory Contiguity in Datacenters Top Picks
• [SOSP'24] FBDetect: Catching Tiny Performance Regressions at Hyperscale through In-Production Monitoring Best Paper
• [OSDI'24] ServiceLab: Preventing Tiny Performance Regressions at Hyperscale through Pre-Production Testing Best Paper
• [OSDI'24] MAST: Global Scheduling of ML Training across Geo-Distributed Datacenters at Hyperscale
• [OSDI'24] Optimizing Resource Allocation in Hyperscale Datacenters: Scalability, Usability, and Experiences
• [SC'24] Accelerating Communication in Deep Learning Recommendation Model Training with Dual-Level Adaptive Lossy Compression
• [ASPLOS'24] Expanding Datacenter Capacity with DVFS Boosting: A safe and scalable deployment experience
• [ICML'24] Wukong: Towards a Scaling Law for Large-Scale Recommendation
• [MLSys'24] Disaggregated Multi-Tower: Topology-aware Modeling Technique for Efficient Large Scale Recommendation
• [WWW'24] Rankitect: Ranking Architecture Search Battling World-class Engineers at Meta Scale
• [NeurIPS'24] CoMERA: Computing- and Memory-Efficient Training via Rank-Adaptive Tensor Optimization
• [NeurIPS'24] Task Me Anything
• [NSDI'24] QuickUpdate: A Real-Time Personalization System for Large-Scale Recommendation Model
• [RecSys'24] Toward 100TB Recommendation Models With Embedding Offloading
• Movie Gen: A Cast of Media Foundation Models
• DCPerf: An open source benchmark suite for hyperscale compute applications
• PyTorch Distributed Shampoo Winning the MLCommon Training Algorithms Competition

2023

• Llama 2: Open Foundation and Fine-Tuned Chat Models
• [ISCA'23] Contiguitas: The Pursuit of Physical Memory Contiguity in Datacenters Best Paper
• [IEEE Micro'23] IOCost: Block IO Control for Containers in Datacenters Top Picks
• [ISCA'23] MTIA: First Generation Silicon Targeting Meta's Recommendation Systems
• [ISCA'23] Shared Microexponents: A Little Shifting Goes a Long Way
• [ISCA'23] TPP: Transparent Page Placement for CXL-Enabled Tiered-Memory
• [ISCA'23] Mystique: Enabling Accurate and Scalable Generation of Production AI Benchmarks
• [MLSys'23] RecD: Deduplication for End-to-End Deep Learning Recommendation Model Training Infrastructure
• [OSDI'23] AdaEmbed: Adaptive Embedding for Large-Scale Recommendation Models
• [ISPASS'23] Characterization of Data Compression in Datacenters
• [ASPLOS'23] Ditto: End-to-End Application Cloning for Networked Cloud Services
• [VLDB'23] PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel
• Microscaling Data Formats for Deep Learning
• FlexShard: Flexible Sharding for Industry-Scale Sequence Recommendation Models

2022

• [ASPLOS'22] TMO: Transparent Memory Offloading in Datacenters Best Paper
• [ASPLOS'22] RecShard: Statistical Feature-Based Memory Optimization for Industry-Scale Neural Recommendation
• [OSDI'22] Unity: Accelerating DNN Training Through Joint Optimization of Algebraic Transformations and Parallelization
• [ISCA'22] Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models
• [NSDI'22] Check-N-Run: a Checkpointing System for Training Deep Learning Recommendation Models
• [ICDCS'22] Supporting Massive DLRM Inference through Software Defined Memory
• [DLP-KDD'22] DHEN: A deep and hierarchical ensemble network for large-scale click-through rate prediction

2021

• [IEEE Micro'21] Understanding Acceleration Opportunities at Hyperscale Top Picks
• [Micro'21] Low-Precision Hardware Architectures Meet Recommendation Model Inference at Scale
• [Big Data'21] Efficient Soft-Error Detection for Low-precision Deep Learning Recommendation Models
• [ATC'21] Improving Performance of Flash Based Key-Value Stores Using Storage Class Memory as a Volatile Memory Extension
• First-Generation Inference Accelerator Deployment at Facebook
• FBGEMM: Enabling High-Performance Low-Precision Deep Learning Inference

2020

• [ASPLOS'20] Accelerometer: Understanding Acceleration Opportunities for Data Center Overheads at Hyperscale
• [DLP-KDD'20] Training Deep Learning Recommendation Model with Quantized Collective Communications
• Mixed-Precision Embedding Using a Cache
• FioSynth: A representative I/O benchmark and data visualizer for data center workloads,
• Accelerometer and SoftSKU: Improving hardware platform performance for diverse microservices

2019

• Deep Learning Recommendation Model for Personalization and Recommendation Systems
• [ISCA'19] SoftSKU: Optimizing Server Architectures for Microservice
• [MLSys'19] Post-Training 4-bit Quantization of Embedding Tables
• [MLSys'19] Bandana: Using Non-Volatile Memory for Storing Deep Learning Models
• [ATC'19] Who's Afraid of Uncorrectable Bit Errors? Online Recovery of Flash Errors with Distributed Redundancy
• A Study of BFLOAT16 for Deep Learning Training