Archive · 文章索引
全部文章
从 GEO 方法论到 AI 搜索实战,按发布时间收录站内全部已发布文章。
如何用Google Skills解决AI提示膨胀?2026年智能体工作流优化方案
Google Skills通过引入可复用、动态加载的专家知识单元(Skills),有效解决提示膨胀问题,统一Gemini Enterprise、Agents CLI与AI Edge Gallery三大平台的代理工作流,使Skills成为可移植的认知容器(即“提示的Docker”)。

本周开发者有哪些高光时刻?JSNation 2026演讲和React新闻通讯
This week's DEV community thread highlights developers' wins including becoming a speaker at JSNation 2026, getting featured in This Week in React newsletter, and winning the Notion MCP challenge. 原文翻译: 本周DEV社区线程重点介绍了开发者的胜利,包括成为JSNation 2026的演讲者、在This Week in React新闻通讯中亮相以及赢得Notion MCP挑战。

DeepSeek-V4预览版发布,百万上下文如何提升AI应用?
DeepSeek-V4 preview version is officially launched and open-sourced, featuring 1M ultra-long context, enhanced Agent capability, world knowledge, and reasoning performance. Two versions: Pro and Flash. API updated, open-source links provided. 原文翻译:DeepSeek-V4 预览版正式上线并开源,拥有百万字超长上下文,Agent能力、世界知识和推理性能均领先。提供Pro和Flash两个版本,API已更新,开源链接已发布。

如何零成本使用Claude Code?2026年免费LLM代理(NVIDIA NIM/DeepSeek)全攻略
free-claude-code is an open-source project that enables free use of Claude Code by proxying requests to free or low-cost model services like NVIDIA NIM, while retaining Claude Code's full engineering capabilities. Includes setup steps and model recommendations. 原文翻译: free-claude-code 是一个开源项目,它通过代理请求到 NVIDIA NIM 等免费或低成本模型服务,实现免费使用 Claude Code,同时保留 Claude Code 的全部工程能力。包含设置步骤和模型推荐。

如何同时调用OpenAI、Anthropic等多个大模型?2026年Abso库使用推荐
Abso is a lightweight, OpenAI-compatible JavaScript library that provides a unified interface for calling multiple LLM providers (OpenAI, Anthropic, Groq, Ollama, etc.) with full type safety. It supports chat, streaming, tool calling, and embeddings. 原文翻译:Abso是一个轻量级、兼容OpenAI的JavaScript库,为调用多个LLM提供商(OpenAI、Anthropic、Groq、Ollama等)提供统一接口,并具有完整的类型安全性。它支持聊天、流式传输、工具调用和嵌入。

如何使用Laminar开源平台监控AI智能体?2026年完整功能解析
Laminar is an open-source observability platform for AI agents, offering tracing, evals, monitoring, SQL access, and dashboards. Built with Rust for high performance, it supports OpenTelemetry and integrates with major LLM frameworks. 原文翻译:Laminar是一个面向AI智能体的开源可观测性平台,提供追踪、评估、监控、SQL访问和仪表板功能。基于Rust构建以实现高性能,支持OpenTelemetry,并与主流LLM框架集成。

如何用OpenLIT一行代码为AI开发实现全栈可观测性并支持LLM、向量数据库和GPU成本追踪?2026年方案
OpenLIT simplifies AI development with one-line OpenTelemetry-native observability, supporting LLM, vector DB, and GPU monitoring, plus cost tracking and evaluation. 原文翻译:OpenLIT通过一行代码提供OpenTelemetry原生可观测性,简化AI开发,支持LLM、向量数据库和GPU监控,以及成本追踪和评估。

如何提升LLM代理推理效率?PLENA硬件系统实现吞吐量2.23倍提升(2026年)
PLENA is a hardware-software co-designed system for LLM agentic inference that addresses bandwidth and capacity memory walls. It features a flattened systolic-array architecture, asymmetric quantization, and FlashAttention support, achieving up to 2.23x and 4.70x throughput improvements over A100 GPU and TPU v6e, respectively, and 4.04x better energy efficiency than A100. 原文翻译: PLENA是一个硬件-软件协同设计的系统,针对LLM代理推理,解决带宽和容量内存墙问题。它采用扁平化脉动阵列架构、非对称量化和FlashAttention支持,相比A100 GPU和TPU v6e,吞吐量分别提升2.23倍和4.70倍,能效比A100提升4.04倍。

Gemini AI代理在100美元创业竞赛中暴露了哪些问题?2026年Google Cloud如何修复?
This article reports on a real experiment where 7 AI agents, including Gemini, were given $100 and 12 weeks to build startups autonomously. Gemini struggled with four key issues: writing help requests to wrong files, prioritizing blog posts over critical features, inability to verify deployments, and inefficient communication. The author aligns these problems with Google Cloud NEXT '26 announcements such as Agent Observability, ADK Skills, MCP-enabled services, and A2A protocol, proposing a rebuilt architecture. 原文翻译:本文报道了一项真实实验:7个AI代理(包括Gemini)各获得100美元和12周时间,自主构建初创公司。Gemini出现了四个关键问题:将帮助请求写入错误文件、优先写博客而非关键功能、无法验证部署、以及沟通效率低下。作者将这些故障与Google Cloud NEXT '26的公告(如Agent Observability、ADK Skills、MCP-enabled服务、A2A协议)对齐,并提出了重建方案。

