Not available in English yet
Anthropic 给 Claude Code 加 83 条指令 — 终端 AI 正在变成开发者的新操作系统
Related Reading
More on #Anthropic
AI PPT Solutions Diverge Wildly — Two 67k-Star GitHub Projects Pick Sides
PPT Master (41k stars) outputs native Office; frontend-slides (26k stars) ships HTML. Two Claude Code Skills, opposite bets on AI's white-collar futur
4 Parallel Agents Beat 1: AI's Winning Play Shifts From Models to Systems
GPT-5.6 defaults to 4 parallel agents; NVIDIA's AVO aces ARC-AGI-3 — August signals say multi-agent is overtaking single-model scaling.
Longer Chats, Dumber AI: Agent Bottleneck Isn't Memory—It's Your PPT
An engineer argues Agents stall because they use chat logs as working memory. When artifacts become directly readable, editable, and verifiable, Agent
Qwen 1-bit is still 6x slower — companies eyeing local LLMs should wait
Qwen's 1-bit quantization runs 6x slower than the 4-bit version with ~70% accuracy. Local LLM deployment isn't ready to replace cloud APIs.
Nemotron's '16GB' Was a Lie—One Dev Proved It, Broke Off-the-Shelf Tools
NVIDIA's Nemotron was secretly faking low-memory versions—one dev audited 443 files, found labels lied. His fix works but breaks LM Studio/Ollama.
Enterprise AI Knowledge Bases Miss the Point — Bug Sits in Retrieval, Not LLM
RAG is now standard for enterprise knowledge bases but keeps misfiring. We trace the fault to retrieval, not the LLM. New tools mark its maturation.