ADK Course #4 - Run your ADK Agents 4 ways - Web App, API Server, CLI, Python Script

In this video, we explore *four powerful ways to run your ADK agents* — whether you're building for web demos, API integrations, or full programmatic control. Learn how each method works and when to use it.

-----------------------------

Udemy Course (get completion certificate, Q&A, bite-sized lectures)
https://www.udemy.com/course/google-a...

-----------------------------

👉 Code used in this video (version_2):

https://theailanguage.com/onlySubscri...

Please subscribe, allow pop-ups, and log in to access our GitHub repos.

-----------------------------

CHAPTERS
00:00 Quick Preview of the four methods
03:46 WEB: adk web
08:19 API: adk api_server
12:43 PROGRAM: agent runner
26:16 CLI: adk run
29:33 Summary

-----------------------------

👇 Here’s a quick summary of the four methods:

1️⃣ *ADK Web*
Command: `adk web ./agents`

Launches a browser-based UI
Best for quick testing, manual debugging, or demos

2️⃣ *ADK API Server*
Command: `adk api_server ./agents`

Starts an HTTP API backend for your agents
Ideal for building custom frontends or integrating with tools via REST

3️⃣ *Programmatic Python Script*
Command: `uv run python3 -m agent_runner`

Run your agent directly using Python code and the ADK SDK
Great for full control, CLI tools, and backend workflows

4️⃣ *ADK CLI Run*
Command: `adk run agents/root_website_builder`

Quickest way to run an agent from the terminal
Perfect for one-off runs and testing without writing code

🛠️ Whether you're building advanced agent pipelines or testing locally, these methods give you the flexibility you need.

📌 Don’t forget to like, comment, and subscribe for more tutorials on building AI agents with Google’s ADK!


#ADK, #API_SERVER, #ADKWEB, #RUNNER, #ADK_CLI, #GoogleADK, #AgentDevelopmentKit, #AIagents, #AItools, #PythonAgents, #OpenAgents, #AIautomation, #LLMapps, #GenAI, #GeminiAgents, #APIAgents, #CLItools, #DeveloperTools, #BuildWithADK