<?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>模型 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E6%A8%A1%E5%9E%8B/</link><description>Recent content in 模型 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>Sat, 27 Jun 2026 15:00:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E6%A8%A1%E5%9E%8B/index.xml" rel="self" type="application/rss+xml"/><item><title>代码生成模型对比选型</title><link>https://guijiagi.com/posts/code-gen-model-comparison/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/code-gen-model-comparison/</guid><description>代码生成模型对比选型</description></item><item><title>多模态模型选型指南</title><link>https://guijiagi.com/posts/multimodal-model-selection-guide/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/multimodal-model-selection-guide/</guid><description>多模态模型选型指南</description></item><item><title>函数调用模型选型</title><link>https://guijiagi.com/posts/function-calling-model-selection/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/function-calling-model-selection/</guid><description>函数调用模型选型</description></item><item><title>模型量化技术原理与实践</title><link>https://guijiagi.com/posts/model-quantization-principles/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/model-quantization-principles/</guid><description>模型量化技术原理与实践</description></item><item><title>轻量级模型部署选型</title><link>https://guijiagi.com/posts/lightweight-model-selection/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/lightweight-model-selection/</guid><description>轻量级模型部署选型</description></item><item><title>数学推理模型能力对比</title><link>https://guijiagi.com/posts/math-reasoning-model-comparison/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/math-reasoning-model-comparison/</guid><description>数学推理模型能力对比</description></item><item><title>长文本处理模型对比</title><link>https://guijiagi.com/posts/long-text-model-comparison/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/long-text-model-comparison/</guid><description>长文本处理模型对比</description></item><item><title>LLM 知识蒸馏：从大模型到小模型的能力迁移</title><link>https://guijiagi.com/posts/llm-knowledge-distillation/</link><pubDate>Thu, 25 Jun 2026 12:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/llm-knowledge-distillation/</guid><description>深入讲解 LLM 知识蒸馏的完整体系：黑盒蒸馏、白盒蒸馏、在线/离线蒸馏，涵盖 Logit 蒸馏、中间层蒸馏、指令蒸馏、响应蒸馏等方法，附 PyTorch 代码实现与效果对比。</description></item><item><title>代码大模型横评：Codex vs Claude Code vs DeepSeek-Coder vs Qwen-Coder</title><link>https://guijiagi.com/posts/code-model-comparison/</link><pubDate>Thu, 25 Jun 2026 12:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/code-model-comparison/</guid><description>深度横评 2026 年四大代码大模型：GPT-5.5-Code、Claude Code、DeepSeek-Coder-V3、Qwen2.5-Coder 在代码生成、调试、重构、代码审查等场景的实战表现。</description></item><item><title>微软 MAI 模型家族：从 1B 到 1T 的全覆盖</title><link>https://guijiagi.com/posts/microsoft-mai-models/</link><pubDate>Thu, 25 Jun 2026 12:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/microsoft-mai-models/</guid><description>微软发布自研 MAI 模型家族，覆盖 1B 到 1T 参数全规模，挑战 OpenAI 依赖，重塑 AI 战略版图。</description></item></channel></rss>