Compare Architectures

Side-by-side: use case, requirements, latency, cost, strengths, weaknesses, failure modes, example — and when to choose each.

FrameworkNo framework
Use caseGraph orchestration, checkpointing, streaming, or many integrations neededSingle agent or workflow with a handful of tools
RequirementsLearn the abstraction; track releases; accept its opinionsProvider SDK, ~200 lines of loop and tool dispatch
LatencyExtra layers; usually negligible, sometimes hidden callsExactly the calls you wrote
CostHidden prompts and retries can add tokensFully visible
StrengthsBatteries included; community patterns; persistence and tracing hooksTotal control, transparent, trivial to debug
WeaknessesLeaky abstractions, churn, hard-to-debug magicYou re-implement checkpointing, streaming, retries
Failure modesBreaking upgrade; unable to see what prompt was sentHome-grown loop lacks limits and observability
ExampleDurable multi-step workflow with human approval and resumeA support bot with 6 tools and a RAG lookup
Choose this whenChoose a framework once you need durable state, graphs, or streaming and can name the problem it solves.Choose no framework first: build the loop yourself, learn the failure modes, then adopt a framework for a specific need.