1Major upgrades to Ray Serve: 88% lower latency and 11.1x higher throughput (opens in new tab)(anyscale.com)1robertnishihara4h ago1
2SkyRL brings Tinker to your GPUs (2025) (opens in new tab)(novasky-ai.notion.site)24robertnishihara1mo ago5
3vLLM large scale serving: DeepSeek 2.2k tok/s/h200 with wide-ep (opens in new tab)(blog.vllm.ai)147robertnishihara2mo ago54
4Massively Parallel Agentic Simulations with Ray (opens in new tab)(anyscale.com)2robertnishihara6mo ago0
5Deploy DeepSeek‑R1 with VLLM and Ray Serve on Kubernetes (opens in new tab)(anyscale.com)1robertnishihara7mo ago0
6An Open Source Stack for AI Compute: Kubernetes and Ray and PyTorch and VLLM (opens in new tab)(anyscale.com)1robertnishihara7mo ago0
9AsyncFlow: An Asynchronous Streaming RL Framework for LLM Post-Training (opens in new tab)(arxiv.org)4robertnishihara8mo ago0
11Large-Scale Deployment of Ray in Tencent's Weixin AI Infrastructure (opens in new tab)(anyscale.com)2robertnishihara8mo ago0
12Uv and Ray: Pain-Free Python Dependencies in Clusters (opens in new tab)(anyscale.com)44robertnishihara9mo ago10
13Roll: Reinforcement Learning Optimization for Large-Scale Learning (opens in new tab)(github.com)1robertnishihara9mo ago0
14An Open Source Stack for AI Compute: Kubernetes and Ray and PyTorch and VLLM (opens in new tab)(anyscale.com)1robertnishihara9mo ago0
15Uv and Ray: Pain-Free Python Dependencies in Clusters (opens in new tab)(anyscale.com)1robertnishihara1y ago0
