AGI Frontier
AI Wiki
围绕持续变化的 AI 主题，汇总定义、近期信号、关键问题与可追溯来源。
主题知识页
14 个主题-
Simon Willison: AI Engineering Notes
A people and opinions page collecting Simon Willison notes on models, tools, agents, and developer workflows。
-
Qwen Model Updates
A topic page for Qwen model releases across language, image, translation, safety, and post-training updates。
-
Open Models Watch
A tracking page for open-weight and open-source model releases, benchmarks, and ecosystem dynamics。
-
Model Context Protocol (MCP)
MCP connects models, tools, data sources, and clients as a protocol layer for agentic systems。
-
Jason Liu: AI Engineering and Consulting
A people and opinions page for Jason Liu writing on context engineering, RAG, structured outputs, and AI consulting。
-
Inference Serving
Inference serving covers latency, throughput, batching, KV cache, routing, quantization, and cost for running models in production。
-
Context Engineering
Context engineering is the practice of designing the information, tools, memory, and state an AI system receives。
-
Coding Agents
Coding agents inspect repositories, edit files, run commands, call tools, and iterate toward software changes。
-
Claude Code and Coding Agents
A product and workflow page for Claude Code, Codex, and adjacent coding-agent systems。
-
AI Safety, Risk, and Governance
AI safety and governance covers frontier risk, monitoring, evaluation, policy, and risk management frameworks。
-
AI Infrastructure
AI infrastructure covers inference, serving, routing, evaluation, observability, RAG, and hardware/data-center dependencies。
-
AI Browsers and Computer Use
AI browsers and computer-use agents operate web pages, desktop apps, and visual interfaces through model-guided actions。
-
AI Agents
AI agents are model-driven systems that plan, call tools, use memory, and iterate across multi-step tasks。
-
Agent Evals and Harnesses
Agent evals measure long-running AI systems through tasks, environments, harnesses, traces, and failure analysis。