ai-dynamo / nixlLinks
NVIDIA Inference Xfer Library (NIXL)
☆729Updated last week
Alternatives and similar repositories for nixl
Users that are interested in nixl are comparing it to the libraries listed below
Sorting:
- Efficient and easy multi-instance LLM serving☆512Updated 2 months ago
- Perplexity GPU Kernels☆531Updated 3 weeks ago
- KV cache store for distributed LLM inference☆368Updated 2 weeks ago
- Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond☆674Updated 3 weeks ago
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆437Updated last week
- Disaggregated serving system for Large Language Models (LLMs).☆737Updated 7 months ago
- OME is a Kubernetes operator for enterprise-grade management and serving of Large Language Models (LLMs)☆322Updated this week
- A low-latency & high-throughput serving engine for LLMs☆450Updated last month
- NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the …☆235Updated last week
- Dynamic Memory Management for Serving LLMs without PagedAttention☆444Updated 6 months ago
- CUDA checkpoint and restore utility☆393Updated 2 months ago
- A tool for bandwidth measurements on NVIDIA GPUs.☆571Updated 7 months ago
- NVIDIA NCCL Tests for Distributed Training☆126Updated 2 weeks ago
- Materials for learning SGLang☆658Updated last week
- GLake: optimizing GPU memory management and IO transmission.☆491Updated 8 months ago
- Byted PyTorch Distributed for Hyperscale Training of LLMs and RLs☆893Updated this week
- ☆324Updated 2 weeks ago
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆234Updated this week
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆122Updated 2 years ago
- Offline optimization of your disaggregated Dynamo graph☆110Updated this week
- torchcomms: a modern PyTorch communications API☆295Updated this week
- Zero Bubble Pipeline Parallelism☆437Updated 6 months ago
- RDMA and SHARP plugins for nccl library☆214Updated last week
- NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs☆621Updated last month
- A validation and profiling tool for AI infrastructure☆352Updated this week
- Distributed Compiler based on Triton for Parallel Systems☆1,251Updated 2 weeks ago
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,182Updated 3 months ago
- Microsoft Collective Communication Library☆374Updated 2 years ago
- A throughput-oriented high-performance serving framework for LLMs☆918Updated last month
- NCCL Profiling Kit☆149Updated last year