Subscribe
Topic

SGLang

SGLang is an open-source engine for serving LLMs fast. Posts compare SGLang with vLLM, TensorRT-LLM and TGI, explain its RadixAttention prefix-tree KV cache, and cover SGLang v0.5.18's faster cold starts.

Articles

1-3
  1. SGLang v0.5.18 Cuts LLM Cold Starts by 2.4x with Overlapped Weight Loading and CUDA Graph Capture

    The open-source LLM serving engine SGLang has released version 0.5.18, introducing an overlapped startup engine that significantly reduces cold-start latency for large language models, alongside communication kernel optimizations and expanded architecture support. Comprising 710 pull requests from 212 contributors, the release addresses operational overheads in LLM infrastructure where autoscaling, rolling cluster deployments, and worker node recovery frequently pay steep restart penalties. O

    AI
    4 min read
  2. Prefix-Tree KV Cache Management in Production: RadixAttention, Tree-Structured LRU Eviction, and Token-Level Sharing in SGLang and vLLM

    Prefix-Tree KV Cache Management in Production: RadixAttention, Tree-Structured LRU Eviction, and Token-Level Sharing in SGLang and vLLM Autoregressive large language model inference is heavily constrained by memory bandwidth and the computational overhead of the prefill phase. For workloads such as multi-turn conversations, autonomous agent tool loops, few-shot prompt evaluations, and tree-search decoding, consecutive requests often share substantial token prefixes. In a standard multi-turn ses

    AI
    7 min read
  3. LLM Serving Frameworks in Production: SGLang vs. vLLM vs. TensorRT-LLM Architecture, Memory Management, and Throughput Trade-Offs

    Serving large language models in production has moved beyond simple iteration-level scheduling. While basic continuous batching solved the fundamental problem of GPU starvation during autoregressive decoding, modern production demands have shifted the engineering bottleneck toward KV cache memory reuse, compilation overhead, and structured generation efficiency. Three dominant serving runtimes currently lead high-throughput deployments: vLLM, SGLang, and NVIDIA TensorRT-LLM. Each framework appr

    Inference
    4 min read