Product
5 Interestana reports mention SGLang, newest first. Each links back to its publisher, with an AI summary of the original reporting where we have found no publisher terms that forbid one.
Tsinghua University and CUHK researchers open-sourced AKV, a serving system for KV cache engineering in LLM agents. The system reduces average cost by 27.7% and increases average task score by 1.7 points across tested models and benchmarks.
Google Cloud has released llm-d 1.0, the first stable version of its open-source, Kubernetes-native distributed inference stack. This layer integrates vLLM or SGLang model servers into a coordinated pool, enhancing self-hosted agent and RL workloads.
DeepSeek has open-sourced DeepInfer, the inference framework behind its DeepSeek-V4.1-Flash model, under the Apache 2.0 license. The release offers a rare look at a frontier lab's serving stack.
The Institute of Foundation Models (IFM), a frontier research lab established by MBZUAI in May 2025, has released K2 Horizon, a comprehensive suite of six open-source large language models. This release, described by IFM as the largest fully open-source model…
Advanced Micro Devices (AMD) has released Instella-MoE-16B-A3B, a fully open Mixture-of-Experts (MoE) large language model trained from scratch on its Instinct MI300X and MI325X Graphics Processing Units (GPUs). This model boasts a total of 16 billion parameters, but crucially, it…