Joe Fernandez
23b740df3b
* fix(deploy): simplifying Agent Engine deployment * revise and update ae deploy doc structure * respond to review comments * docs: break up Models page into multiple pages - Create page per model and access method - Push Models topic up to top level TOC entry * updates based on review feedback * fix broken links due to page move * consistency updates and review input updates * update gemini-2.0-flash to 2.5 * respond to review comments * docs: update Python API docs to 1.22.0 (#1129) * build(deps): bump golang.org/x/crypto (#987) Bumps the go_modules group with 1 update in the /tools/feature-matrix directory: [golang.org/x/crypto](https://github.com/golang/crypto). Updates `golang.org/x/crypto` from 0.42.0 to 0.45.0 - [Commits](https://github.com/golang/crypto/compare/v0.42.0...v0.45.0) --- updated-dependencies: - dependency-name: golang.org/x/crypto dependency-version: 0.45.0 dependency-type: indirect dependency-group: go_modules ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Sita Lakshmi Sangameswaran <sitalakshmi@google.com> * Update model version from gemini-2.0 to gemini-2.5, since gemini-2.0-flash will be deprecated on Feb 5,2026. (#1132) Co-authored-by: Joe Fernandez <joefernandez@users.noreply.github.com> * Changes model MODEL_2_0_FLASH to MODEL_2_5_FLASH as gemini-2.0 is deprecating (#1131) * Update model constants to use GEMINI 2.5 * Corrected model name --------- Co-authored-by: Joe Fernandez <joefernandez@users.noreply.github.com> * Add Agent Runtime pages for Web Interface and Command Line; reorganize existing Agent Runtime pages (#1130) * Add pages for adk web, adk run, reorganize runtime pages * Minor edit * Updates based on review * docs: break up Models page into multiple pages - Create page per model and access method - Push Models topic up to top level TOC entry * minor update * add placeholder file --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: Polong Lin <polong-lin@users.noreply.github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Sita Lakshmi Sangameswaran <sitalakshmi@google.com> Co-authored-by: llalitkumarrr <llalitkumarrr@google.com> Co-authored-by: Dinesh Thumma <160909147+DineshThumma9@users.noreply.github.com> Co-authored-by: Kristopher Overholt <koverholt@google.com>
4.7 KiB
Agents
In Agent Development Kit (ADK), an Agent is a self-contained execution unit designed to act autonomously to achieve specific goals. Agents can perform tasks, interact with users, utilize external tools, and coordinate with other agents.
The foundation for all agents in ADK is the BaseAgent class. It serves as the fundamental blueprint. To create functional agents, you typically extend BaseAgent in one of three main ways, catering to different needs – from intelligent reasoning to structured process control.
Core Agent Categories
ADK provides distinct agent categories to build sophisticated applications:
-
LLM Agents (
LlmAgent,Agent): These agents utilize Large Language Models (LLMs) as their core engine to understand natural language, reason, plan, generate responses, and dynamically decide how to proceed or which tools to use, making them ideal for flexible, language-centric tasks. Learn more about LLM Agents... -
Workflow Agents (
SequentialAgent,ParallelAgent,LoopAgent): These specialized agents control the execution flow of other agents in predefined, deterministic patterns (sequence, parallel, or loop) without using an LLM for the flow control itself, perfect for structured processes needing predictable execution. Explore Workflow Agents... -
Custom Agents: Created by extending
BaseAgentdirectly, these agents allow you to implement unique operational logic, specific control flows, or specialized integrations not covered by the standard types, catering to highly tailored application requirements. Discover how to build Custom Agents...
Choosing the Right Agent Type
The following table provides a high-level comparison to help distinguish between the agent types. As you explore each type in more detail in the subsequent sections, these distinctions will become clearer.
| Feature | LLM Agent (LlmAgent) |
Workflow Agent | Custom Agent (BaseAgent subclass) |
|---|---|---|---|
| Primary Function | Reasoning, Generation, Tool Use | Controlling Agent Execution Flow | Implementing Unique Logic/Integrations |
| Core Engine | Large Language Model (LLM) | Predefined Logic (Sequence, Parallel, Loop) | Custom Code |
| Determinism | Non-deterministic (Flexible) | Deterministic (Predictable) | Can be either, based on implementation |
| Primary Use | Language tasks, Dynamic decisions | Structured processes, Orchestration | Tailored requirements, Specific workflows |
Agents Working Together: Multi-Agent Systems
While each agent type serves a distinct purpose, the true power often comes from combining them. Complex applications frequently employ multi-agent architectures where:
- LLM Agents handle intelligent, language-based task execution.
- Workflow Agents manage the overall process flow using standard patterns.
- Custom Agents provide specialized capabilities or rules needed for unique integrations.
Understanding these core types is the first step toward building sophisticated, capable AI applications with ADK.
What's Next?
Now that you have an overview of the different agent types available in ADK, dive deeper into how they work and how to use them effectively:
- LLM Agents: Explore how to configure agents powered by large language models, including setting instructions, providing tools, and enabling advanced features like planning and code execution.
- Workflow Agents: Learn how to orchestrate tasks using
SequentialAgent,ParallelAgent, andLoopAgentfor structured and predictable processes. - Custom Agents: Discover the principles of extending
BaseAgentto build agents with unique logic and integrations tailored to your specific needs. - Multi-Agents: Understand how to combine different agent types to create sophisticated, collaborative systems capable of tackling complex problems.
- Models: Learn about the different LLM integrations available and how to select the right model for your agents.