62 related articles

A Bilibili creator ran Qwen 122B with 256K context on just 8GB VRAM + 64GB RAM using llama.cpp. Full breakdown of quantization, deployment params, performance, and cost-effective alternatives.

LLM JSON output unstable in your Agent? This guide covers 6 engineering layers: constrained decoding, validation retry, fake tool calls, Logit Masking, Schema contracts, and anti-pattern locking.
The Hidden Costs of AI Coding Tools: A…
Claude Code consumes up to 33,000 tokens per request before reading your prompt, vs. 7,000 for OpenCode. We break down why this 5x gap matters for cost, latency, and context windows.

In-depth hands-on review of Zhipu AI's flagship GLM-5.2: a 1M-token context window and API pricing just one-fifth of GPT/Claude. Covers website building, Chrome extensions, 3D game cloning, and agentic workflows.

Real debugging case: when 400MB of source code and 40K files caused an infinite crash loop, MiniMax M3, DeepSeek, and Hunyuan all gave wrong answers. GPT-4.1 mini found the root cause after an hour of deep reasoning.

An in-depth hands-on review of Zhipu AI's flagship GLM-5.2: 1M-token context, strong coding, mature agent workflows—at one-fifth the price of top frontier models. Covers website testing, Cursor integration, MCP tooling, and production migration.

OpenClaw author Peter shares his thinking on AI coding: when compute is no longer the bottleneck, developer attention becomes the new ceiling. Learn three practical skills—Agent Transcript, Auto-Review, and the Crapbox sandbox.

A deep dive into Claude Code: the difference between Terminal and Device Agents, enterprise selection advice, and how to use Claude Code with DeepSeek in China.

GitHub Trending July 6: Agent skill ecosystem explodes with taste-skill, marketingskills, dotnet/skills; multi-Agent orchestration matures; privacy-first projects thrive.

GitHub Trending July 5: Claude Code Skill ecosystem explodes, AI pen-testing tool Strix gains +2137 Stars, and local-first privacy apps surge.

Deep analysis of why Chinese AI models (DeepSeek, Qwen) need opinionated tool frameworks to work properly, exploring training differences, function calling maturity, and framework selection advice.

CLI-WeChat-Bridge is an open-source tool that bridges AI CLI tools like Codex, Claude Code, and Open Code to WeChat, supporting voice input, file transfer, and multi-CLI parallel switching.

A detailed walkthrough of Claude Code + Opus 4.7 in action: from CCSwitch setup to model configuration, completing a complex payment feature in 4 hours for $60 with zero hand-written code.

In-depth review of OpenCode, an open-source AI coding assistant. Covers its three-layer architecture, setup, building a to-do app, and model comparisons with DeepSeek Flash and more.

Complete guide to Claude Code + Opus setup via CCSwitch for unified model management. Real-world case: payment system built in 4 hours with zero hand-written code, covering installation, Prompt engineering, and model selection.

Claude-mem is an open-source AI memory tool that gives Claude Code, Codex, and other AI coding assistants cross-session memory via semantic compression and vector retrieval — just 50 tokens of overhead, fully local storage.

Deep analysis of oh-my-openagent plugin's critical flaws: hardcoded Claude Opus 4.7 identity misleads non-Claude users, prompt injection doubles token costs. Includes alternatives and developer tips.

8 free terminal tools—Bat, Eza, Chafa, Zoxide, TLDR, Miru, Yazi, LazyGit—to dramatically boost your Claude Code and AI coding workflow efficiency.
The Complete Guide to Claude Code: Cho…
A deep dive into Claude Code's core advantages as a terminal AI programming tool, how it differs from Device Agents, and practical setup steps to get started with enterprise-grade AI coding.

In-depth review of OpenCode, a free open-source AI coding assistant. Covers installation, features, and cost comparison with Claude Code to help developers decide if this zero-config alternative is worth switching to.