Liang Wu 048deeaeb7 feat(live): Handle input transcription differently for Gemini Live 3.1 models
For Gemini 3.1 Flash Live, input transcriptions are now marked as finished in the first response containing text, rather than being accumulated as partial updates. This aligns with the expected behavior of the 3.1 API.

Co-authored-by: Liang Wu <wuliang@google.com>
PiperOrigin-RevId: 929514971
2026-06-09 18:22:29 -07:00
2026-06-04 15:29:20 -07:00
2025-11-03 13:33:53 -08:00

Agent Development Kit (ADK) 2.0

License

An open-source, code-first Python framework for building, evaluating, and deploying sophisticated AI agents with flexibility and control.


⚠️ BREAKING CHANGES FROM 1.x

This release includes breaking changes to the agent API, event model, and session schema. Sessions generated by ADK 2.0 are readable by ADK 1.28+ (extra fields will be ignored), but are incompatible with older 1.x versions.


🔥 What's New in 2.0

  • Workflow Runtime: A graph-based execution engine for composing deterministic execution flows for agentic apps, with support for routing, fan-out/fan-in, loops, retry, state management, dynamic nodes, human-in-the-loop, and nested workflows.

  • Task API: Structured agent-to-agent delegation with multi-turn task mode, single-turn controlled output, mixed delegation patterns, human-in-the-loop, and task agents as workflow nodes.

🚀 Installation

pip install google-adk

Requirements: Python 3.11+.

To install optional integrations, you can use the following command:

pip install "google-adk[extensions]"

The release cadence is roughly bi-weekly.

Quick Start

Agent

from google.adk import Agent

root_agent = Agent(
    name="greeting_agent",
    model="gemini-2.5-flash",
    instruction="You are a helpful assistant. Greet the user warmly.",
)

Workflow

from google.adk import Agent, Workflow

generate_fruit_agent = Agent(
    name="generate_fruit_agent",
    instruction="Return the name of a random fruit. Return only the name.",
)

generate_benefit_agent = Agent(
    name="generate_benefit_agent",
    instruction="Tell me a health benefit about the specified fruit.",
)

root_agent = Workflow(
    name="root_agent",
    edges=[("START", generate_fruit_agent, generate_benefit_agent)],
)

Run Locally

# Interactive CLI
adk run path/to/my_agent

# Web UI (supports multi-agent directories or pointing directly to a single agent folder)
adk web path/to/agents_dir

📚 Documentation

🤝 Contributing

See CONTRIBUTING.md for details.

📄 License

This project is licensed under the Apache 2.0 License — see the LICENSE file for details.

S
Description
An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.|GitHub 镜像 21.3k · 🍴 3.9k
https://github.com/google/adk-python Readme Apache-2.0 79 MiB
Languages
Python 77.5%
JavaScript 20.7%
Jupyter Notebook 1.3%
HTML 0.3%