Anthropic plugin

The genkit-plugin-anthropic package provides access to Anthropic’s Claude models through the Genkit framework.

Installation

uv add genkit-plugin-anthropic

Configuration

from genkit import Genkit
from genkit.plugins.anthropic import Anthropic, anthropic_name

ai = Genkit(
    plugins=[Anthropic()],
    model=anthropic_name('claude-3-5-haiku'),
)

Authentication

Set the ANTHROPIC_API_KEY environment variable or pass it directly:

ai = Genkit(
    plugins=[Anthropic(api_key='your-api-key')],
)

Available Models

• claude-3-5-haiku - Fast and efficient
• claude-sonnet-4 - Balanced performance
• claude-sonnet-4-5 - Latest with thinking/reasoning support

Basic Usage

response = await ai.generate(
    prompt='Explain quantum computing in simple terms.',
)
print(response.text)

Structured Output

from pydantic import BaseModel, Field

class Character(BaseModel):
    name: str = Field(description='Character name')
    backstory: str = Field(description='Character backstory')
    abilities: list[str] = Field(description='List of abilities')

response = await ai.generate(
    prompt='Generate a fantasy RPG character',
    output_schema=Character,
)
print(response.output)  # Character instance

Thinking and Reasoning

Claude Sonnet 4.5 and later models support extended thinking:

response = await ai.generate(
    model=anthropic_name('claude-sonnet-4-5'),
    prompt='Solve this logic puzzle step by step...',
    config={
        'thinking': {'type': 'enabled', 'budget_tokens': 1024},
        'max_output_tokens': 4096,
    },
)
print(response.text)

Tool Calling

from pydantic import BaseModel, Field

class WeatherInput(BaseModel):
    location: str = Field(description='City name')

@ai.tool()
def get_weather(input: WeatherInput) -> str:
    """Get current weather for a location."""
    return f'72°F and sunny in {input.location}'

response = await ai.generate(
    prompt='What is the weather in San Francisco?',
    tools=['get_weather'],
)
print(response.text)

Streaming

from genkit import ActionRunContext

@ai.flow()
async def streaming_story(topic: str, ctx: ActionRunContext) -> str:
    stream_response = ai.generate_stream(
        prompt=f'Write a short story about {topic}',
    )
    async for chunk in stream_response.stream:
        ctx.send_chunk(chunk.text)
    return (await stream_response.response).text

Multimodal (Image Understanding)

from genkit import Part, TextPart, MediaPart, Media

response = await ai.generate(
    prompt=[
        Part(root=TextPart(text='Describe this image')),
        Part(root=MediaPart(media=Media(
            url='https://example.com/image.jpg',
            content_type='image/jpeg',
        ))),
    ],
)
print(response.text)

Configuration Options

from genkit import ModelConfig

response = await ai.generate(
    prompt='Your prompt here',
    config=ModelConfig(
        temperature=0.7,
        max_output_tokens=1000,
    ),
)