bytedance / InfiniStoreLinks
KV cache store for distributed LLM inference
☆371Updated last month
Alternatives and similar repositories for InfiniStore
Users that are interested in InfiniStore are comparing it to the libraries listed below
Sorting:
- Efficient and easy multi-instance LLM serving☆517Updated 3 months ago
- NVIDIA Inference Xfer Library (NIXL)☆753Updated this week
- ☆328Updated last month
- Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond☆701Updated 2 weeks ago
- Perplexity GPU Kernels☆536Updated last month
- Fast OS-level support for GPU checkpoint and restore☆260Updated 2 months ago
- GLake: optimizing GPU memory management and IO transmission.☆491Updated 8 months ago
- High performance Transformer implementation in C++.☆143Updated 10 months ago
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆118Updated 6 months ago
- Offline optimization of your disaggregated Dynamo graph☆121Updated this week
- A low-latency & high-throughput serving engine for LLMs☆454Updated last month
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆236Updated 2 weeks ago
- Dynamic Memory Management for Serving LLMs without PagedAttention☆446Updated 6 months ago
- DeepSeek-V3/R1 inference performance simulator☆169Updated 8 months ago
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆442Updated this week
- Materials for learning SGLang☆682Updated last week
- torchcomms: a modern PyTorch communications API☆302Updated this week
- Fast and memory-efficient exact attention☆104Updated this week
- Venus Collective Communication Library, supported by SII and Infrawaves.☆123Updated this week
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆407Updated last month
- Disaggregated serving system for Large Language Models (LLMs).☆747Updated 8 months ago
- Allow torch tensor memory to be released and resumed later☆184Updated last week
- A lightweight design for computation-communication overlap.☆194Updated 2 months ago
- Stateful LLM Serving☆89Updated 9 months ago
- ☆73Updated 11 months ago
- Research prototype of PRISM — a cost-efficient multi-LLM serving system with flexible time- and space-based GPU sharing.☆49Updated 3 months ago
- OME is a Kubernetes operator for enterprise-grade management and serving of Large Language Models (LLMs)☆324Updated last week
- The driver for LMCache core to run in vLLM☆59Updated 10 months ago
- ☆134Updated last week
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆70Updated 2 weeks ago