<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>推理引擎 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E6%8E%A8%E7%90%86%E5%BC%95%E6%93%8E/</link><description>Recent content in 推理引擎 on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Thu, 16 Jul 2026 11:39:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E6%8E%A8%E7%90%86%E5%BC%95%E6%93%8E/index.xml" rel="self" type="application/rss+xml"/><item><title>大模型推理引擎横评：vLLM、SGLang、TensorRT-LLM</title><link>https://guijiagi.com/posts/b1-f50aea03/</link><pubDate>Thu, 16 Jul 2026 11:39:00 +0800</pubDate><guid>https://guijiagi.com/posts/b1-f50aea03/</guid><description>对比分析三大主流推理引擎的架构设计、性能特征与适用场景，给出选型建议</description></item><item><title>大模型推理部署方案对比：vLLM、SGLang与TensorRT-LLM</title><link>https://guijiagi.com/posts/b2-e0dd1517/</link><pubDate>Thu, 16 Jul 2026 10:38:00 +0800</pubDate><guid>https://guijiagi.com/posts/b2-e0dd1517/</guid><description>深度对比2026年三大主流大模型推理引擎的架构设计、性能表现与适用场景</description></item><item><title>vLLM 2026社区进展：高性能推理引擎的进化</title><link>https://guijiagi.com/posts/vllm-2026-community/</link><pubDate>Thu, 02 Jul 2026 11:26:00 +0800</pubDate><guid>https://guijiagi.com/posts/vllm-2026-community/</guid><description>2026年vLLM社区发展动态与技术创新，LLM推理引擎的标杆</description></item><item><title>Hermes Agent 爱马仕智能体技术架构深度解析</title><link>https://guijiagi.com/posts/hermes-agent-architecture/</link><pubDate>Sun, 28 Jun 2026 11:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/hermes-agent-architecture/</guid><description>深度解析 Hermes Agent 爱马仕智能体的技术架构设计、核心组件、推理引擎与应用实践</description></item><item><title>MLX：Apple Silicon 上的大模型推理框架</title><link>https://guijiagi.com/posts/mlx-apple-silicon-inference/</link><pubDate>Sun, 28 Jun 2026 11:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/mlx-apple-silicon-inference/</guid><description>深度解析 Apple MLX 框架在 2026 年的大模型推理能力，涵盖性能优化、模型适配与实际部署实践</description></item><item><title>SGLang 2026：结构化生成的高性能推理引擎</title><link>https://guijiagi.com/posts/sglang-2026-inference-engine/</link><pubDate>Sun, 28 Jun 2026 11:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/sglang-2026-inference-engine/</guid><description>深入探索 SGLang 2026 版本的 RadixAttention、结构化输出、推理优化与生产部署实践</description></item><item><title>TGI（Text Generation Inference）2026 指南</title><link>https://guijiagi.com/posts/tgi-2026-guide/</link><pubDate>Sun, 28 Jun 2026 11:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/tgi-2026-guide/</guid><description>全面介绍 HuggingFace TGI 2026 版本的架构、部署、优化与生产实践，含与 vLLM 的深度对比</description></item><item><title>vLLM 2026 生产部署完全指南</title><link>https://guijiagi.com/posts/vllm-2026-deployment-guide/</link><pubDate>Sun, 28 Jun 2026 11:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/vllm-2026-deployment-guide/</guid><description>从单机部署到分布式集群，全面覆盖 vLLM 2026 的安装、配置、优化与生产运维实践</description></item><item><title>vLLM 部署深度指南：高吞吐 LLM 推理引擎</title><link>https://guijiagi.com/posts/vllm-deployment-deep/</link><pubDate>Thu, 25 Jun 2026 10:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/vllm-deployment-deep/</guid><description>深入解析 vLLM 架构、PagedAttention 原理、Continuous Batching、量化支持与分布式推理部署。</description></item><item><title>SGLang 掐理引擎指南：超越 vLLM 的新选择</title><link>https://guijiagi.com/posts/sglang-inference-engine/</link><pubDate>Wed, 24 Jun 2026 14:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/sglang-inference-engine/</guid><description>SGLang 原理详解、与 vLLM/TensorRT-LLM 性能对比、安装使用、适用场景分析</description></item></channel></rss>