vllm-project / routerLinks
A high-performance and light-weight router for vLLM large scale deployment
☆82Updated 3 weeks ago
Alternatives and similar repositories for router
Users that are interested in router are comparing it to the libraries listed below
Sorting:
- Genai-bench is a powerful benchmark tool designed for comprehensive token-level performance evaluation of large language model (LLM) serv…☆251Updated this week
- ArcticInference: vLLM plugin for high-throughput, low-latency inference☆375Updated this week
- ☆340Updated 2 weeks ago
- Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond☆753Updated last week
- Efficient and easy multi-instance LLM serving☆521Updated 4 months ago
- Dynamic Memory Management for Serving LLMs without PagedAttention☆457Updated 7 months ago
- Allow torch tensor memory to be released and resumed later☆202Updated last week
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆626Updated last week
- A low-latency & high-throughput serving engine for LLMs☆467Updated last week
- A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM☆198Updated this week
- KV cache store for distributed LLM inference☆385Updated 2 months ago
- Tile-Based Runtime for Ultra-Low-Latency LLM Inference☆527Updated 3 weeks ago
- Offline optimization of your disaggregated Dynamo graph☆151Updated this week
- A lightweight design for computation-communication overlap.☆213Updated 3 weeks ago
- Perplexity GPU Kernels☆553Updated 2 months ago
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆86Updated last month
- torchcomms: a modern PyTorch communications API☆320Updated last week
- Toolchain built around the Megatron-LM for Distributed Training☆80Updated last month
- Accelerating MoE with IO and Tile-aware Optimizations☆542Updated last week
- Utility scripts for PyTorch (e.g. Make Perfetto show some disappearing kernels, Memory profiler that understands more low-level allocatio…☆80Updated 4 months ago
- Open Model Engine (OME) — Kubernetes operator for LLM serving, GPU scheduling, and model lifecycle management. Works with SGLang, vLLM, T…☆356Updated this week
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆123Updated 3 weeks ago
- High performance Transformer implementation in C++.☆148Updated last year
- Materials for learning SGLang☆717Updated 2 weeks ago
- ☆130Updated last year
- Perplexity open source garden for inference technology☆332Updated 3 weeks ago
- LLM Serving Performance Evaluation Harness☆82Updated 10 months ago
- ☆147Updated last year
- Stateful LLM Serving☆94Updated 10 months ago
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆275Updated 3 months ago