facebookresearch / param
PArametrized Recommendation and Ai Model benchmark is a repository for development of numerous uBenchmarks as well as end to end nets for evaluation of training and inference platforms.
☆124Updated this week
Related projects ⓘ
Alternatives and complementary repositories for param
- NCCL Profiling Kit☆113Updated 4 months ago
- Synthesizer for optimal collective communication algorithms☆98Updated 7 months ago
- RDMA and SHARP plugins for nccl library☆162Updated last week
- Microsoft Collective Communication Library☆54Updated this week
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆250Updated this week
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆112Updated last year
- ☆73Updated last year
- Repository for MLCommons Chakra schema and tools☆39Updated 10 months ago
- RCCL Performance Benchmark Tests☆50Updated 3 weeks ago
- This is a plugin which lets EC2 developers use libfabric as network provider while running NCCL applications.☆147Updated this week
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆57Updated 6 months ago
- ROCm Communication Collectives Library (RCCL)☆270Updated this week
- Microsoft Collective Communication Library☆322Updated last year
- Repository for MLCommons Chakra schema and tools☆67Updated last week
- oneAPI Collective Communications Library (oneCCL)☆206Updated this week
- An Efficient Pipelined Data Parallel Approach for Training Large Model☆70Updated 3 years ago
- TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches☆64Updated last year
- ☆47Updated last year
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 23)☆79Updated last year
- ☆33Updated 2 months ago
- An experimental parallel training platform☆52Updated 7 months ago
- Fine-grained GPU sharing primitives☆140Updated 4 years ago
- 🔮 Execution time predictions for deep neural network training iterations across different GPUs.☆56Updated last year
- Research and development for optimizing transformers☆125Updated 3 years ago
- Pytorch process group third-party plugin for UCC☆20Updated 7 months ago
- FTPipe and related pipeline model parallelism research.☆41Updated last year
- An interference-aware scheduler for fine-grained GPU sharing☆112Updated 6 months ago
- Magnum IO community repo☆79Updated 6 months ago
- oneCCL Bindings for Pytorch*☆86Updated 3 weeks ago
- Chimera: Efficiently Training Large-Scale Neural Networks with Bidirectional Pipelines.☆46Updated 11 months ago