🍥

记录并分享日常

Section

708 pages

Posts

Godot Starter Workflow: Build 2D Games with VS Code and the Codex Extension

GLM 5.2 Goes Open Source: Million-Token Context, Agent Coding, and the Cost of Local Deployment

Codex CLI OSS Mode: Set Up Ollama, LM Studio, OpenRouter, and Local Models

Three Ways Codex Can Use a Computer: Computer Use, the Chrome Extension, and the In-App Browser

OpenAI API Basics: Writing Chat Completions, Function Calls, and Streaming Output

How to Apply for the AI Version of Alipay Abao Beta: Invite Codes, Entry Points, and Notes

WD PR4100 / PR2100 Serial Console: UART Pins, Wiring, and Risk Notes

Why Is AI Coding Still Unstable? Loop Engineering May Be the Next Step

MultiTalk: Audio-Driven Multi-Person Conversational Video Generation

InfiniteTalk: Audio-Driven Talking Video Generation With Long-Video Support

Windows 11 Low Latency Profile: Using Short CPU Boosts To Improve System Responsiveness

The Biggest Trap in Vibe Coding: Requirement Drift

addyosmani/agent-skills Guide: Engineering Workflows for AI Coding Agents

iptv-org/iptv Tutorial: GitHub M3U Playlist, Channel Lists, and EPG Entry Points

Anthropic Suspends Fable 5 and Mythos 5 Over Security Concerns

Anthropic Responds to US Government Directive Suspending Access to Fable 5 and Mythos 5

gd-agentic-skills: Turning Godot Development Experience into AI-Callable Skills

GodotPrompter: Giving AI Coding Assistants More Godot Project Experience

Browser Cache and Cloudflare Cache Are Not the Same: Cache-Control, Edge TTL, and Publish Refresh

Holo 3.1 Local Deployment Guide: Run a Computer Use Agent with llama.cpp and OpenClaw

GPT-5.6 Rumor Explained: 1.5M Context Window, Iris-Alpha, and GPT-5.5 Comparison

SpaceX-API: An Open-Source Space Data Interface Maintained by Fans

OpenAI's Reported Ona Deal: Codex Is Moving From Coding Assistant to Cloud Agent Platform

Dario Amodei's New Essay: AI Is Moving Too Fast for Regulation, Jobs, and Global Competition to Keep Up

CLAUDE-FABLE-5 System Prompt Analysis: Safety Rules, Tools, Search, and Limits

Gemma 4 MTP Tuning: Pushing Toward 120 tokens/s With an assistant Draft Model

What Is Gemma 4 assistant-MTP: How Multi-Token Prediction Draft Models Speed Up Inference

Running Gemma 4 12B on 8GB VRAM: Tuning llama-cli Hybrid Offload Parameters

Deploying DiffusionGemma Locally: Running Google’s Text Diffusion Model with vLLM

DiffusionGemma: Google Brings Diffusion Models into Text Generation

How to Choose an AI Memory System: Mem0, Letta, Zep, Cognee, and Memobase Compared

Can AI Read a Motherboard? WD PR2100 HDMI, UART, and Backplane Interface Analysis

Codex Hooks Guide: Privacy Checks, Tool Review, Logging, and Validation

Leiden Declaration: How Mathematicians Are Responding to AI in Research

How to Handle the HTTP/2 Bomb Vulnerability: Impact and Mitigation for CVE-2026-49975

OpenTalking vs LongCat-Video: One for Real-Time Conversation, One for High-Quality Digital Human Video

What Is OpenTalking? An Open-Source Framework for Getting AI Digital Human Conversations Running

MeTube: Add a Browser Download Panel to yt-dlp

Using Claude Fable 5 for Investment Analysis: Research Notes, Bear Cases, and Risk Lists

Open-LLM-VTuber Guide: Run a Local AI Companion with Live2D, Voice, and LLMs

What Is turbovec? A Rust Vector Index That Saves Memory for Local RAG

Probabilistic TRM: A Little Randomness Makes a Tiny Reasoning Model Much Stronger

SpaceX AI1 Satellite Revealed: Moving AI Data Centers Into Orbit Sounds Bold, but the Hard Parts Are Real

Antimalware Service Executable High CPU Usage? Don’t Rush to Disable Defender

Is Apple Finally Taking AI Seriously? WWDC26 Siri AI and Apple Intelligence Explained

Loops Replace Prompts: Loop Engineering Is Changing How AI Agents Work

Codex Pricing and Usage Limits: Token Rate Card Explained

Claude Usage Limits: 5-Hour Resets, Weekly Caps, and Claude Code

Claude Fable 5 Prompting Guide: Long Tasks, Agents, and Effort Settings

Reading the Claude Fable 5 Product Page: Built for Long Tasks, Agents, and Hard Coding Work