ModelRefs / Canonical Architecture Patterns — Library
Canonical Architecture Patterns — Library
The canonical architecture-pattern registry: RAG, agentic, fine-tuning, multimodal, realtime, batch, edge, and hybrid stacks with layers, data flow, scaling profiles, and safety controls.
What this reference supports
Canonical Architecture Patterns — Library: This hub organizes related ModelRefs references into a crawlable starting point. Use it to narrow the problem, identify relevant profiles or guides, and continue into detailed evidence and implementation material.
Canonical Architecture Patterns — Library: Items are connected across models, providers, benchmarks, workflows, tools, and guides. Those relationships explain where an option fits, what it depends on, and which adjacent decisions still need validation.
Canonical Architecture Patterns — Library: Catalogue presence is not an endorsement or universal ranking. Compare candidates against your own requirements and review each page's sources, freshness notes, limitations, and coverage gaps.
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Canonical Architecture Patterns — Library.
Frequently asked questions
What is a canonical architecture pattern?
A canonical architecture pattern is a reusable, model-agnostic recipe for assembling a full LLM/AI system — its layers, runtime topology, safety controls, scaling profile, and measurable effectiveness.
What is RAG?
Retrieval-Augmented Generation (RAG) is an architecture that grounds LLM outputs in retrieved documents, reducing hallucination and enabling knowledge freshness without retraining.
When should I use an agentic architecture?
Use agentic patterns when the task requires multi-step reasoning, tool calls, memory, or dynamic planning — such as code generation workflows, research agents, or customer-support automation.
What topology should I choose?
Start with a monolith for development speed. Move to pipeline or microservices as you need to scale individual components independently. Event-driven is best for high-concurrency async workloads.
How is architecture effectiveness scored?
Effectiveness = weighted combination of adoption evidence, citation authority, deployment case studies, benchmark coverage, and ecosystem integration breadth.