Articles

This article provides a comprehensive first-hand account of OpenAI's new GPT-5 model family, based on two weeks of preview access. The author details GPT-5's core characteristics, including its hybrid nature in ChatGPT and simpler API variants (regular, mini, nano) with adjustable reasoning levels. Key specifications like large token limits (272,000 input, 128,000 output) and multimodal input capabilities are highlighted. A significant focus is placed on GPT-5's aggressive pricing strategy, positioning it competitively against other leading models like Claude and Gemini, with a detailed comparison table. Insights from the GPT-5 system card reveal substantial improvements in reducing hallucinations, enhancing instruction following, and minimizing sycophancy, alongside the introduction of 'safe-completions' for nuanced safety responses. The article also critically examines the ongoing challenge of prompt injection, despite GPT-5 showing better resistance than predecessors. Finally, it explores API features such as 'thinking traces' and the 'reasoning_effort' option, concluding with practical examples like SVG generation benchmarks.

This article presents an early-access review of OpenAI's GPT-5, asserting it represents the closest step yet towards Artificial General Intelligence (AGI). The author redefines AGI in the context of LLMs as the ability to 'think with tools,' rather than merely using them. GPT-5 excels in this paradigm, demonstrating advanced capabilities in parallel tool calling and handling complex software engineering tasks like resolving dependency conflicts and building production-ready web applications from a single prompt. The article emphasizes a shift in interaction, suggesting users should 'prompt an agent' by providing a 'compass' of clear, structured guidance rather than extensive context. While highlighting its unparalleled performance in coding and agentic workflows, the review also candidly notes GPT-5's inferiority in creative writing compared to its predecessors.

The article details Zhipu's latest open-source GLM-4.5V multimodal visual reasoning model, which is based on the GLM-4.5 base and has won SOTA in 41 out of 42 public leaderboards, making it the strongest 100B-level open-source multimodal model. Through multiple real-world case studies such as GeoGuessr (guessing addresses from images), Qingming Shanghe Map Grounding (Grounding of Along the River During the Qingming Festival), video-to-frontend code conversion, spatial relationship understanding, UI-to-Code, image recognition, and object counting, the article comprehensively demonstrates GLM-4.5V's excellent capabilities in image, video, and document understanding, especially its untrained "video-to-code" capability, which reflects strong generalization. At the technical level, the article elaborates on the AIMv2-Huge Visual Encoder, MLP adapter, 3D-RoPE, and three-stage (pre-training, SFT, RL) training strategies used by GLM-4.5V. In addition, the article also mentions the model's cost-effective API calls and free resource packs, aiming to lower the barrier for developers to use and promote multimodal AI from 'proof of concept' to 'large-scale deployment.'

Based on insights from MCP co-creator David Soria Parra, this article details the Model-Client Protocol's (MCP) capabilities and future direction. It clarifies that MCP extends beyond simple tool calling to enable richer human-computer interaction. The article introduces MCP's five primitives: Prompt (user-initiated templates), Resource (raw data exposed to the client), Tool (model-driven action calls), Sampling (server requests client for completion), and Roots (client environment information). These primitives—Prompt, Resource, and Tool—form a complete AI application interaction chain. Furthermore, the article emphasizes MCP's future web integration, addressing authentication (OAuth 2.1) and scalability (streaming HTTP mode) solutions. Finally, it previews upcoming features like asynchronous tasks, user interaction requests, an official registry, and multi-modal capabilities, emphasizing MCP's evolution into a system protocol for building rich LLM interaction experiences.

The article provides an in-depth analysis of the GPT-5 launch event, noting its overall lackluster performance and the absence of a transformative upgrade like its predecessors. Benchmark data shows only a slight lead. While GPT-5 demonstrates significant improvements in programming ability, a substantial reduction in hallucination rate, and enhanced context handling, its API pricing strategy, significantly lower than major competitors, is considered a game changer. However, the launch event itself faced criticism due to misleading charts and uninspiring demonstrations, leading to a PR backlash for OpenAI. The article suggests that GPT-5's incremental progress and OpenAI's shift towards price competition signal the end of the AI industry's rapid growth phase, with a transition to a more pragmatic and competitive stage. This implies a need for new breakthroughs in the AI field.

This article, based on Anthropic's 'Building effective agents' and incorporating the author's practical experience, comprehensively analyzes the core concepts and construction strategies of Workflow and Agent in Large Language Model (LLM) applications. The article first distinguishes the applicable scenarios and advantages/disadvantages of Workflow and Agent, and emphasizes the principle of prioritizing simple solutions in LLM application development. Subsequently, it introduces various Agent system design patterns such as enhanced LLMs, Prompt Chaining, Routing, Parallelization, Orchestrator-Worker, and Evaluation-Optimization, and elaborates on the application and combination of these patterns through rich real-world examples (such as content moderation, medical research assistant, literary translation system). In particular, the article emphasizes the critical role of Tool Prompt Engineering in Agent systems, providing practical suggestions for tool design, interface optimization, and continuous improvement. Finally, combined with successful cases of AI-powered Customer Service and Code Generation Agent, it summarizes the core principles and implementation key points for building efficient Agent systems, aiming to provide actionable insights for technology managers, developers, and product managers.

The article details the real feelings and experiences of a senior developer '喵神 (Meow God)' using Claude Code (CC) intensively for one and a half months. The author introduces the concept of 'Vibe coding,' pointing out that AI greatly improves the speed of development iteration, but also brings new challenges such as increased competition, model performance degradation, and resource limitations. The article compares the differences between traditional editor AI and the command-line tool CC, emphasizing CC's advantages in global understanding and forced reliance on AI. The author deeply analyzes CC's strengths and weaknesses, such as its excellent performance in understanding and summarizing tasks, and its limitations in precise refactoring and niche languages. The article discusses the applicable scenarios for 'planning first' and 'practice first' development models, and strongly recommends adopting a 'small step iteration' strategy in most cases. In addition, the author shares practical tips for dealing with AI context limitations, effectively utilizing commands and peripheral tools (such as MCP, voice input), and calls on developers to remain aware and not be constrained by tools.

This article delves into how the Alibaba team, when faced with a core frontend component called 'Specification Panel' that had a large codebase (over 3000 lines), numerous platform adaptation branches, and complex logic, successfully completed a full refactoring to the ICE Architecture in just 10 days by introducing the AI development tool Cursor and combining it with the Claude Model, achieving cross-platform reusability for Web and Weex. The article points out that although the initial code generated by AI tools had issues, the effectiveness of AI was fully realized. This was achieved by establishing and continuously iterating detailed coding standards, Weex platform adaptation rules, and a comprehensive testing system. Practice has proven that AI can not only assist in code generation and greatly improve the production efficiency of test cases (generating 600+ cases), but also achieve 'VibeCoding,' a development approach using natural language-based instructions, through multiple rounds of dialogue. The author proposes a new development paradigm of 'Requirement Delivery = Automatically Maintained Development Rules + Detailed Technical System Analysis and Design Documentation + Comprehensive Test Cases', emphasizing that the probabilistic reasoning nature of large models determine that they need to be under clear constraints and manual verification to efficiently solve complex problems, ultimately achieving a significant result of a 30% increase in R&D efficiency.

This article delves into the memory challenges faced by AI Agents in long-conversation scenarios, particularly the early information loss and resource consumption caused by Large Language Model (LLM) context length limitations. To address these challenges, the article systematically analyzes eight mainstream AI memory strategies, covering basic strategies like 'Full Memory' and 'Sliding Window,' and more advanced strategies like 'Relevance Filtering,' 'Summarization/Compression,' 'Vector Database,' 'Knowledge Graph,' 'Hierarchical Memory,' and 'OS-like Memory Management.' Each strategy is explained in detail, including its core principles, specific characteristics (advantages and disadvantages), and best-suited application scenarios, supplemented by concise simulation code examples to help readers understand its implementation mechanisms. The article emphasizes that developers should flexibly select and combine different memory schemes based on the AI Agent's specific application requirements and system resource constraints, to build more efficient, accurate, and intelligent systems with long-term context awareness.

The author shares their practical experience and methodologies from four months of pair programming with the AI programming assistant Cursor. The article emphasizes that the key to AI collaboration lies in clear requirement descriptions and development plans, and proposes establishing 'rules' to regulate AI behavior, reducing ineffective communication and operational risks. The article details several MCP (Model Context Protocol) tools, including `mcp-feedback-enhanced` for closed-loop feedback, `sequential-thinking` for structured thinking, and `mcp_better_tapd_server` for automated task recording. Through a real-world case study, it demonstrates how to efficiently understand new project code by combining rules and MCP tools. The author concludes that AI tools not only improve efficiency but, more importantly, encourage developers to optimize their thinking patterns, fostering personal growth.