For AI builders
Track coding agents, model APIs, local LLM stacks, RAG workflows, and benchmark signals before choosing what to build with.
Global AI field guide for people who love AI
Follow the models, agents, tools, and experiments that actually matter.
zglg.work is an English-first AI intelligence hub for global readers. It connects daily AI news, model rankings, hands-on field notes, tool guides, and local LLM practice in one practical reading path.
Daily
AI signal
12
English field notes
4
Core hubs
Tools
Guides and calculators
Start Here
Four entry points for global AI readers
Daily AI Signal
A fast reading surface for model launches, AI agents, coding tools, open-source projects, and infrastructure shifts.
Open sectionModel Intelligence
Composite rankings and practical interpretation across Arena, Artificial Analysis, Vals-style tasks, HELM, and real usage.
Open sectionHands-on Field Notes
Original experiments with models and tools, with screenshots, workflows, and honest notes about what actually worked.
Open sectionAI Tool Lab
Tool guides, local deployment notes, calculators, model selectors, and practical setup paths for builders.
Open sectionOriginal Field Notes
Screenshot-heavy AI experiments from the archive
Real task test24 visuals2026-06-02
I Finally Found Free Tokens for MyClaw: Qwen3.6-35B Works Directly (Tutorial)
Hi, I am Guozhen. A lot of readers have told me the same thing: running MyClaw-style agent work burns tokens very quickly. Say one sentence and 100,000 tokens m...
Read field note
2026-05-31 · 29 visuals
I Tested Qwen3.7-Max Against Claude Opus 4.8 and DeepSeek-V4. The Result Surprised Me.
Hi, I am Guozhen. Recently, two new models were released: Qwen3.7-Max and Claude Opus 4.8. On the benchmark list I was watching, Claude Opus 4.8 ranked first. Q...
2026-05-28 · 27 visuals
I Tested a Chinese AI Agent: In 20 Minutes It Produced Word, PPT, Charts, and a Full Report
Hi, I am Guozhen. In the past, if I needed to read several PDF reports, summarize them into a Word document, create a few charts, and then turn everything into...
2026-05-27 · 26 visuals
I Tested a Local AI Agent for One-Person Company Workflows
Hi, I am Guozhen. Some readers told me they are interested in building a "one-person company," but they do not know where to start. In this article, I want to t...
2026-05-26 · 19 visuals
I Tested a Local Knowledge Base That Learns Hundreds of Files and Exports Reports
Hi, I am Guozhen. Many people have a large amount of material on their computers: PDFs, Word documents, Excel files, meeting notes, project documents, papers, c...
For curious readers
Use the daily brief and field notes to follow what changed in AI without getting buried under generic tech noise.
For model evaluators
Compare public benchmark signals with practical constraints: speed, cost, context window, reliability, and deployment fit.
Quick Navigation
Explore by what you want to do today
Skip the vague AI hype. Pick a path below and jump straight into news, model choices, tool workflows, or real experiments.
5-minute catch-upModel decisionsHands-on tests
Catch up on AI today
Model launches, agents, coding tools, and infrastructure moves worth your time.
Open daily briefChoose a model
Compare capability, speed, cost, context, and practical fit before you build.
View rankingsFind an AI tool
Search guides, alternatives, calculators, and workflow recipes by real task.
Open tool workbenchRead real experiments
Screenshot-heavy field notes about tools, models, agents, and local LLM setups.
Browse field notes