Sequential a fixed pipeline
Shape Agents/steps run one after another; each output feeds the next.
Use when The task is a repeatable pipeline with clear, ordered stages (e.g. extract β transform β summarize).
Google Cloud's guide Choose a design pattern for your agentic AI system organizes multi-agent designs around one question: who decides the workflow path? When your code decides, the flow is predetermined and predictable; when the model decides, it's dynamic and adaptive.
Every pattern is a way to arrange one or more agents. The organizing axis is control: does a developer wire the sequence, or does an LLM route at runtime? Start on the deterministic side and move toward autonomy only as the problem demands it.
| Predetermined (code decides) | Dynamic (AI decides) | |
|---|---|---|
| Control | High, repeatable | Lower, adaptive |
| Best for | Well-understood, structured tasks | Ambiguous, open-ended tasks |
| Patterns | Sequential Β· Parallel Β· Loop | Coordinator Β· Hierarchical Β· Swarm |
Shape Agents/steps run one after another; each output feeds the next.
Use when The task is a repeatable pipeline with clear, ordered stages (e.g. extract β transform β summarize).
Shape Independent agents run concurrently; a step merges their results.
Use when Subtasks are independent (speed) or you want diverse perspectives gathered at once.
Shape Repeat a step until an exit condition is met (a quality bar, a max count).
Use when Output improves with iteration and you can define "good enough" (e.g. refine until a score passes).
Here the routing is chosen at runtime by a model. The three shapes differ in who holds authority β a central router, a manager hierarchy, or a flat peer group:
Shape A central agent classifies each request and routes it to the right specialist agent.
Use when Inputs vary and are best handled by different specialists (adaptive routing).
Shape A top orchestrator decomposes the goal and delegates to sub-agents (sometimes via mid-level managers), then integrates results.
Use when The problem is open-ended and benefits from a manager coordinating specialized workers.
Shape No central boss β peer agents collaborate and hand off control to whichever is best suited next.
Use when Complex problems benefit from emergent collaboration or debate among equals.
These apply on top of either side of the axis:
One capable agent with tools handles a moderately complex task β no delegation. The simplest starting point.
A generator agent paired with a critic that validates output β for accuracy and compliance (a.k.a. reflection).
Insert human approval at high-stakes steps β works with any pattern for safety and control.
| If the task is⦠| Pattern | Route |
|---|---|---|
| Moderate, single-scope, tool-using | Single-agent | β |
| A fixed, ordered pipeline | Sequential | Predetermined |
| Independent subtasks / many perspectives | Parallel | Predetermined |
| Iterative refinement to a condition | Loop | Predetermined |
| Varied inputs needing adaptive routing | Coordinator | Dynamic |
| Open-ended, decompose-and-delegate | Hierarchical | Dynamic |
| Emergent collaboration / debate | Swarm | Dynamic |
| High-stakes / needs approvals | + Human-in-the-loop | Any |
| Question | Short answer |
|---|---|
| The organizing question? | Who decides the route β your code (predetermined) or the model (dynamic)? |
| Predetermined patterns? | Sequential, Parallel, Loop. |
| Dynamic patterns? | Coordinator (router), Hierarchical (orchestratorβworkers), Swarm (peers). |
| Swarm vs Hierarchical? | Swarm has no central boss; hierarchical has an orchestrator/managers. |
| Overlays? | Single-agent, review & critique, human-in-the-loop. |
| Core advice? | Prefer the most deterministic pattern that works. |
Six questions on Google's agent design patterns. Pick an answer for each, then check your score.
The other taxonomy β workflows vs agents and the agent loop.
Build these patterns β Google ADK's workflow agents map directly onto them.
Scaffolds for every pattern β coordinator, hierarchical, swarm & more.
The hub β LLMs, agents, protocols, cloud & more.