Global AI Learning Map
Learn AI by following models, agents, tools, and real experiments.
This is the English front door for zglg.work. It is designed for global AI enthusiasts who want a practical map of large models, coding agents, RAG, local LLMs, benchmarks, AI tools, and field-tested workflows.
Daily signal
Today AI Brief
Start with the daily signal layer for model launches, agents, coding tools, open-source updates, and AI infrastructure.
Model choice
Model Benchmark Hub
Read model leaderboards with context instead of chasing a single first-place score.
Tool choice
AI Tool Decision Hub
Compare coding agents, AI search, knowledge tools, local LLM setups, image tools, and practical calculators.
Real tests
Hands-on Field Notes
Read screenshot-rich experiments with real models and tools, preserved as evidence instead of generic summaries.
Learning Tracks
Six tracks for understanding modern AI
Model literacy
Understand context windows, reasoning behavior, multimodal ability, speed, cost, and when smaller models are enough.
AI agents and coding workflows
Move from chat-style prompting to repo-aware agents, tool calls, traces, tests, and reviewable changes.
Local LLMs and private AI
Learn when to use local models, how to think about GPU memory, quantization, data privacy, and offline workflows.
RAG and knowledge systems
Turn documents into useful AI workflows by understanding chunking, embeddings, retrieval, reranking, and evaluation.
Prompting and workflow design
Write prompts as reusable work instructions with inputs, constraints, examples, output formats, and review steps.
AI product judgment
Separate demos from durable products by checking use cases, integration cost, reliability, data policy, and user fit.
Suggested Route
A practical reading path
Follow the daily signal
Use the daily AI brief to see what changed in models, agents, tools, open source, and infrastructure.
Map the model landscape
Read benchmark sources with methodology context, then choose models by scenario rather than by brand.
Pick practical tools
Use tool guides and calculators to compare coding agents, research tools, local setups, and automation stacks.
Study real experiments
Use field notes to see screenshots, prompts, workflow constraints, and results from hands-on AI testing.
Go deep in the archive
Use the Chinese course archive as a deeper source library when you need long-form tutorials and implementation details.
Need deeper source material?
The original archive includes 172 long-form Chinese course collections. Use it when you want extra screenshots, step-by-step implementation notes, or older experiments beyond the English map.
Open original archiveDeepSeek and local model practice
Local LLM setup, model usage, and practical DeepSeek learning materials.
Open archive pathAI agents
Agent concepts, tool use, workflows, and practical implementation notes.
Open archive pathDify and knowledge-base workflows
RAG, knowledge bases, workflow automation, and application building.
Open archive pathGenerative AI foundations
Core AI concepts, model behavior, data workflows, and beginner-friendly foundations.
Open archive pathChoose your next move
If you only have a few minutes, start with the daily brief. If you are choosing a model or tool, use the benchmark hub and workbench. If you want evidence, read the field notes.