Download
DLRM (Deep Learning Recommendation Model) is Meta’s open-source reference implementation for large-scale recommendation systems built to handle extremely high-dimensional sparse features and embedding tables. The architecture combines dense (MLP) and sparse (embedding) branches, then interacts features via dot product or feature interactions before passing through further dense layers to predict click-through, ranking scores, or conversion probabilities. The implementation is optimized for performance at scale, supporting multi-GPU and multi-node execution, quantization, embedding partitioning, and pipelined I/O to feed huge embeddings efficiently. It includes data loaders for standard benchmarks (like Criteo), training scripts, evaluation tools, and capabilities like mixed precision, gradient compression, and memory fusion to maximize throughput.
Features
- Hybrid architecture combining sparse embeddings and dense MLP branches
- Efficient feature interaction (e.g. dot product, permutation) between sparse and dense features
- Multi-GPU and distributed training with embedding partitioning and gradient synchronization
- Support for quantization, memory optimization, and pipelined embedding I/O
- Training / evaluation support for large-scale datasets like Criteo, Avazu
- Baseline reference for industry and academic recommendation models
Project Samples
