# PydanticAI — Type-Safe AI Agent Framework

> Build production-grade AI agents with type safety, structured outputs, and multi-model support. By the creators of Pydantic and FastAPI.

## Install

Save as a script file and run:

## Quick Use

```bash
pip install pydantic-ai
```

```python
from pydantic_ai import Agent

agent = Agent(
    'anthropic:claude-sonnet-4-6',
    instructions='Be concise, reply with one sentence.',
)
result = agent.run_sync('Where does "hello world" come from?')
print(result.output)
```

---

## Intro

PydanticAI brings the FastAPI developer experience to AI agent building — type-safe, production-grade, and model-agnostic. Built by the Pydantic team (whose validation library powers OpenAI SDK, Anthropic SDK, LangChain, and more). Supports Claude, GPT, Gemini, DeepSeek, and dozens of other providers with structured outputs, tool definitions, and dependency injection. 15,900+ stars.

**Works with**: Claude Code, GitHub Copilot, Gemini CLI, OpenAI Codex, Ollama
---

## Key Features

- **Type safety** — Full IDE auto-completion and static type checking for agents, tools, and outputs
- **Model-agnostic** — OpenAI, Anthropic, Gemini, DeepSeek, Grok, Ollama, and custom models
- **Structured outputs** — Define output schemas with Pydantic models, validated automatically
- **Dependency injection** — Clean separation of business logic and AI configuration
- **MCP support** — Native Model Context Protocol integration
- **Streaming** — Continuous structured output with immediate validation
- **Human-in-the-loop** — Flag tool calls for approval before execution
- **Observability** — Built-in Pydantic Logfire integration for debugging and cost tracking

## Example: Bank Support Agent

```python
from dataclasses import dataclass
from pydantic import BaseModel, Field
from pydantic_ai import Agent, RunContext

@dataclass
class SupportDeps:
    customer_id: int
    db: DatabaseConn

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

support_agent = Agent(
    'openai:gpt-4o',
    deps_type=SupportDeps,
    output_type=SupportOutput,
    instructions='You are a bank support agent.',
)

@support_agent.tool
async def customer_balance(
    ctx: RunContext[SupportDeps], include_pending: bool
) -> float:
    return await ctx.deps.db.customer_balance(
        id=ctx.deps.customer_id,
        include_pending=include_pending,
    )
```

## With Thinking & Web Search

```python
from pydantic_ai.capabilities import Thinking, WebSearch

agent = Agent(
    'anthropic:claude-sonnet-4-6',
    capabilities=[Thinking(), WebSearch()],
)
result = agent.run_sync('What was the largest meteorite found this year?')
```

---



### FAQ

**Q: What is PydanticAI?**
A: Build production-grade AI agents with type safety, structured outputs, and multi-model support. By the creators of Pydantic and FastAPI.

**Q: How do I install PydanticAI?**
A: Check the Quick Use section above for step-by-step installation instructions. Most assets can be set up in under 2 minutes.

## Source & Thanks

> Created by [Pydantic](https://github.com/pydantic). Licensed under MIT.
> [pydantic-ai](https://github.com/pydantic/pydantic-ai) — ⭐ 15,900+
> Docs: [ai.pydantic.dev](https://ai.pydantic.dev)

Thanks to the Pydantic team for bringing type-safe, production-grade tooling to AI agent development.

---
Source: https://tokrepo.com/en/workflows/0313bf39-8bbe-4a50-9445-e5ee8e7280fe
Author: Script Depot