pydantic-ai README

The project's README from github.com/pydantic/pydantic-ai. As one markdown file: README.md.

Pydantic AI

How Python does AI

CI Coverage PyPI versions license Join Slack

Agents, realtime voice, image generation, embeddings. Every model, every interface, typed end to end.

Pydantic AI is the Python AI SDK: a typed, extensible agent loop with every model a string swap away. The same agent runs everywhere you need it: behind a web frontend, in the terminal, on a voice call, on a durable background queue, or as a plain object you call run() on. Image generation and embeddings come in the same box.

Pydantic AI Harness has everything an agent needs for complex, long-running work, snapped on as capabilities, from memory, sub-agents, and context management to a complete coding agent.

View the complete documentation at pydantic.dev/docs/ai.

What are you building?

From simple typed data extraction to complex, long-running multi-agent collaboration, Pydantic AI and Pydantic AI Harness have got you covered.

Coding agent

A complete coding agent in your terminal: workspace-rooted file access, allowlisted shell, repo orientation, planning, and context management that survives long sessions. Here with web search and a second-opinion advisor snapped on alongside:

uv add pydantic-ai pydantic-ai-harness

from pydantic_ai import Agent
from pydantic_ai.capabilities import WebSearch
from pydantic_ai_harness import Advisor, Coder

agent = Agent(
    'anthropic:claude-fable-5',
    capabilities=[
        Coder(),  # files, shell, repo context, planning, sub-agents, context management
        WebSearch(),  # look up docs and error messages on the web
        Advisor('openai:gpt-5.6-sol'),  # a second opinion from another model when stuck
    ],
)
agent.to_cli_sync()

Coder is a regular combined capability, not a black box: use it whole, or use the blocks it bundles directly; the two are equivalent:

capabilities = [
    FileSystem('.'), Shell(cwd='.'), RepoContext(), Planning(), SubAgents(...),
    ClearToolResults(), WarnNearLimits(), ToolOutputLimits(),
]

Run the file and you're chatting with the agent in your terminal. To try it before writing any code, run the exported coder_agent with clai (the Pydantic AI CLI), via uvx:

uvx --with pydantic-ai-harness clai -a pydantic_ai_harness.coder:coder_agent -m anthropic:claude-fable-5

Build this → Coder, from the Harness

Data extraction

Give the agent an output type and tools, and every run comes back validated and typed:

uv add pydantic-ai

from typing import Literal

from pydantic import BaseModel, Field

from pydantic_ai import Agent, RunContext


class Sentiment(BaseModel):
    label: Literal['positive', 'negative', 'neutral']
    score: float = Field(ge=-1, le=1)


agent = Agent('openai:gpt-5.6-sol', output_type=Sentiment)


@agent.tool
def recent_reviews(ctx: RunContext[None], product: str) -> list[str]:
    """Fetch recent review snippets for a product."""
    return ['The new release fixed everything I complained about!']


result = agent.run_sync('How are people feeling about the Extract app?')
print(result.output)
#> label='positive' score=0.9

The @agent.tool function receives a RunContext that carries your dependencies in; the rest of its signature and its docstring become the tool schema, arguments are validated before your code runs, and the run is guaranteed to return a Sentiment, so your IDE, type checker, and the LLM all agree on the returned type.

Build this → Agents, Function Tools, and Structured Output

Durable workflow

Attach TemporalDurability and the same agent runs inside a Temporal workflow under durable execution: every model and tool call becomes a durable activity, so a run working through a background queue survives restarts, failures, and long waits:

uv add "pydantic-ai[temporal]"

from temporalio import workflow

from pydantic_ai import Agent
from pydantic_ai.capabilities import WebFetch, WebSearch
from pydantic_ai.durable_exec.temporal import PydanticAIWorkflow, TemporalDurability

agent = Agent(
    'openai:gpt-5.6-sol',
    instructions='Research the topic and write a structured brief.',
    name='researcher',
    capabilities=[WebSearch(), WebFetch(), TemporalDurability()],
)


@workflow.defn
class ResearchWorkflow(PydanticAIWorkflow):
    __pydantic_ai_agents__ = [agent]

    @workflow.run
    async def run(self, topic: str) -> str:
        result = await agent.run(f'Write a brief on: {topic}')
        return result.output

DBOS and Prefect attach the same way, first-party and co-maintained, with Restate, Kitaru, and Airflow integrations besides.

Build this → Durable Execution

Realtime voice

Put the same agent on a live voice session, tools and capabilities included:

uv add "pydantic-ai[openai-realtime]"

import asyncio

from pydantic_ai import Agent
from pydantic_ai.capabilities import MCP

agent = Agent(
    instructions='You are a helpful voice assistant.',
    capabilities=[MCP('https://internal.example.com/mcp')],  # capabilities work in voice too
)

@agent.tool_plain
def order_status(order_id: str) -> str:
    """Look up the status of an order."""
    return f'Order {order_id}: shipped, arriving Thursday.'

async with agent.realtime('openai:gpt-realtime-2.1').session() as session:
    microphone = asyncio.create_task(session.send_audio(microphone_chunks()))  # your microphone → the model
    speaker = asyncio.create_task(play_audio(session.stream_audio()))  # model audio → your speaker
    async for part in session.stream_transcripts():
        print(f'{part.speaker}: {part.transcript}')

The model calls your tools mid-conversation while it keeps talking, and every session is instrumented; voice is just another frontend, on OpenAI Realtime, Gemini Live, Azure, and xAI Grok Voice.

Build this → Realtime Voice

Image generation

Generate an image with a dedicated image model, no agent run required:

uv add pydantic-ai

from pathlib import Path

from pydantic_ai import ImageGenerator

generator = ImageGenerator('openai:gpt-image-2')
result = generator.generate_sync('A minimalist logo for a coffee shop called Extract.')
Path('logo.png').write_bytes(result.image.data)

That standalone image API is for when your application decides; when an agent run decides, there is provider-native generation with output_type=BinaryImage for a typed image output, and the ImageGeneration capability with its fallbacks for models that generate no images of their own.

Build this → Image Generation

Why Pydantic AI

Built by the Pydantic team: Pydantic Validation is the validation layer of the OpenAI SDK, the Anthropic SDK, the Google ADK, LangChain, and most of the AI ecosystem (and the foundation FastAPI was built on). Pydantic AI brings that same feeling to agents.

Putting it together: a bank support agent

A typed support agent showing several features working together: dependency injection, function tools, structured output, a reusable capability bundling the customer context, and an on-demand capability the model loads only when the conversation calls for it:

from dataclasses import dataclass

from pydantic import BaseModel, Field

from pydantic_ai import Agent, Capability, RunContext

from bank_database import DatabaseConn


@dataclass
class SupportDependencies:  # inject any client: DB pools, HTTP APIs, user info
    customer_id: int
    db: DatabaseConn


class SupportOutput(BaseModel):
    support_advice: str = Field(description='Advice returned to the customer')
    block_card: bool = Field(description="Whether to block the customer's card")
    risk: int = Field(description='Risk level of query', ge=0, le=10)


customer_context = Capability[SupportDependencies](  # a reusable unit of tools + instructions
    id='customer-context',
    description="Who the customer is and what's on their account.",
)


@customer_context.instructions
async def add_customer_name(ctx: RunContext[SupportDependencies]) -> str:
    customer_name = await ctx.deps.db.customer_name(id=ctx.deps.customer_id)
    return f"The customer's name is {customer_name!r}"


@customer_context.tool  # signature and docstring become the tool schema the LLM sees
async def customer_balance(
    ctx: RunContext[SupportDependencies], include_pending: bool
) -> float:
    """Returns the customer's current account balance."""
    return await ctx.deps.db.customer_balance(
        id=ctx.deps.customer_id,
        include_pending=include_pending,
    )


refunds = Capability[SupportDependencies](  # deferred: loads on demand, like a skill
    id='refunds',
    description='Refund eligibility and refund status.',
    defer_loading=True,
)


@refunds.tool
async def refund_status(ctx: RunContext[SupportDependencies]) -> str:
    """Look up the refund status for the customer's most recent charge."""
    return await ctx.deps.db.refund_status(id=ctx.deps.customer_id)


support_agent = Agent(
    'openai:gpt-5.6-sol',
    deps_type=SupportDependencies,
    output_type=SupportOutput,  # the run returns a validated SupportOutput, typed as such
    instructions=(
        'You are a support agent in our bank, give the '
        'customer support and judge the risk level of their query.'
    ),
    capabilities=[customer_context, refunds],
)


...  # in a real use case: more tools, longer instructions


async def main():
    deps = SupportDependencies(customer_id=123, db=DatabaseConn())
    result = await support_agent.run('What is my balance?', deps=deps)
    print(result.output)
    """
    support_advice='Hello John, your current account balance, including pending transactions, is $123.45.' block_card=False risk=1
    """

    result = await support_agent.run('I just lost my card!', deps=deps)
    print(result.output)
    """
    support_advice="I'm sorry to hear that, John. We are temporarily blocking your card to prevent unauthorized transactions." block_card=True risk=8
    """

    result = await support_agent.run(  # the model loads `refunds` on demand, then answers
        'Was I refunded for the duplicate charge on my last statement?', deps=deps
    )
    print(result.output)
    """
    support_advice='Good news, John: the duplicate charge on your last statement was refunded on 2026-05-01.' block_card=False risk=1
    """

For the annotated walkthrough and Logfire tracing, see the same example in the docs.

Next Steps

Part of the Pydantic Stack

Everything you need to ship production-grade AI agents:

pydantic-ai (pydantic/pydantic-ai)

Searching Database…

20.7K ·GitHub
Pydantic-ai
What is this?Pydantic AI is the Python AI SDK for building agents: a typed, extensible agent loop where every model is a string swap away. The same agent runs behind a web frontend, in the terminal, on a voice call, or as a plain object you call run() on.
Is this a framework or a library — what do I get when I install it?It's a Python SDK you import and code against: agents, realtime voice, image generation, and embeddings come in the same box. An optional Harness adds snap-on capabilities, and agents can even be defined in YAML/JSON specs instead of code.
How do I get it running — do I need an API key first?No — there's a built-in 'test' model, so you can start without any API key. Install the package and define an agent in a few lines of Python; to try an agent before writing any code, run the bundled Coder agent with clai (the Pydantic AI CLI) via uvx.
Which models and providers can I plug in?Virtually every provider — OpenAI, Anthropic, Google, Bedrock, Azure AI Foundry, Groq, Mistral, xAI, Ollama, and dozens more — swapped by changing a string. The Pydantic AI Gateway adds one key for all of them, with failover and cost monitoring built in.
Anything experimental or rough around the edges I should know about?Only one piece is flagged: ACP, which serves your agent as an editor agent, is experimental. Everything else — CLI, web chat, realtime voice frontends — is presented as standard.
What does 'typed end to end' actually mean in practice?Structured outputs, typed dependency injection, and typed tools: your IDE, type checker, and the LLM all agree on what an agent returns, so whole classes of errors move from runtime to write-time. Tool arguments are validated before your code runs, and Pydantic Graph extends the same typing to graph-based workflows.
What is the Harness, and what counts as a 'capability'?The Harness is the add-on layer for complex, long-running work: memory, sub-agents, context management, even a complete coding agent, all snapped on as capabilities. A capability bundles tools, instructions, hooks, and model settings into one reusable unit — core ships MCP and web search, and ready agents like Coder and Researcher are just composed capabilities.
Can one agent do realtime voice and image generation too?Yes. Voice is just another frontend: the model calls your tools mid-conversation while it keeps talking, on OpenAI Realtime, Gemini Live, Azure, and xAI Grok Voice. For images there's a standalone API, provider-native generation via a typed output, and an ImageGeneration capability with fallbacks for models that can't generate images themselves.
How do I debug, monitor, and test what my agents do?Instrumentation is OpenTelemetry-native and works with any OTel backend; one line lights up Pydantic Logfire for real-time debugging, tracing, and cost tracking. Pydantic Evals tests agent behavior the way pytest tests code, and durable engines like Temporal keep long runs alive across restarts and failures.
What are similar projects?hive, voltagent, harness-sdk, parlant, open-multi-agent, cayu, SystemOneHarness, gitagent
Which topics is it in?pydantic, genai, harness-engineering, validation, api-gateway, voice, embeddings, harness, agent-framework, evaluation, image-generation, observability
Creator
pydantic
Size
313.1 MB
Stars
20.7K
Updated
CommitOct 18Releasev2.57.0Oct 16
Age
2y
Commits · 90 Days