====== PydanticAI ======

**PydanticAI** is a type-safe Python agent framework built by the creators of Pydantic. With **15.7K GitHub stars** and 390+ contributors, it brings the "FastAPI feeling" to GenAI development — validated outputs, dependency injection, structured tool contracts, and observability as first-class concerns rather than afterthoughts.

{{tag>framework python type-safe agents pydantic dependency-injection}}

===== Overview =====

PydanticAI was created by the Pydantic team (Samuel Colvin et al.) after they found that despite virtually every Python AI framework using Pydantic for validation, none provided the ergonomic developer experience that FastAPI brought to web development. Released in mid-2024, PydanticAI v1.0 shipped in September 2025 and has rapidly iterated to v1.70+ by March 2026. The framework validates agent inputs, outputs, tool parameters, and dependency types at every boundary using Pydantic models, surfacing errors at development time rather than in production.

===== Key Features =====

  * **Type Safety** — All agent inputs, outputs, tools, and dependencies validated by Pydantic models
  * **Dependency Injection** — Clean, testable architecture with typed dependencies passed to tools and instructions
  * **Structured Outputs** — LLM responses validated against Pydantic schemas, ensuring reliable data extraction
  * **Model Agnostic** — Supports 15+ providers: OpenAI, Anthropic, Gemini, DeepSeek, Grok, Cohere, Mistral, Ollama
  * **Tool Calling** — Type-safe function tools with automatic parameter validation
  * **PydanticAI Graph** — Graph-based workflows for complex multi-step agent pipelines
  * **Pydantic Evals** — Built-in evaluation framework for testing agent behavior
  * **MCP & A2A Support** — Model Context Protocol and Agent-to-Agent protocol integration
  * **Logfire Integration** — OpenTelemetry-native observability via Pydantic Logfire
  * **Capabilities** — Reusable bundles of tools, hooks, instructions, and model settings

===== Architecture =====

PydanticAI follows a clean, layered architecture centered on type-safe boundaries:

<mermaid>
graph TD
    subgraph Agent["Agent - Deps, Output"]
        A[Instructions: System Prompt]
        B[Tools: Validated Functions]
        C[Output Schema: Pydantic Model]
    end
    Agent --> D[Model Interface Layer]
    D --> E[OpenAI]
    D --> F[Anthropic]
    D --> G[Gemini / Ollama / Others]
    D --> H[Observability: Logfire]
    H --> I[Traces / Spans / Metrics / OpenTelemetry]
</mermaid>

===== Code Example =====

A type-safe agent with dependency injection and structured output:

<code python>
from pydantic import BaseModel
from pydantic_ai import Agent

class CityInfo(BaseModel):
    name: str
    country: str
    population: int
    known_for: list[str]

# Create a type-safe agent with structured output
agent = Agent(
    "openai:gpt-4o",
    output_type=CityInfo,
    system_prompt="You are a geography expert. Return accurate city data.",
)

# Run the agent — output is validated against CityInfo schema
result = agent.run_sync("Tell me about Paris")
print(f"City: {result.output.name}")
print(f"Country: {result.output.country}")
print(f"Population: {result.output.population:,}")
print(f"Known for: {', '.join(result.output.known_for)}")

# Type-safe tool with dependency injection
from dataclasses import dataclass

@dataclass
class WeatherDeps:
    api_key: str

weather_agent = Agent(
    "openai:gpt-4o",
    deps_type=WeatherDeps,
    system_prompt="Use the weather tool to answer questions.",
)

@weather_agent.tool
async def get_weather(ctx, city: str) -> str:
    api_key = ctx.deps.api_key  # Type-safe access
    return f"Weather in {city}: 22C, sunny"
</code>

===== Packages =====

The PydanticAI ecosystem consists of focused packages:

^ Package ^ Purpose ^
| **pydantic-ai** | Core agent framework with tools and DI |
| **pydantic-ai-slim** | Minimal install without provider extras |
| **pydantic-graph** | Graph-based multi-step workflows |
| **pydantic-evals** | Agent evaluation and testing framework |

===== References =====

  * [[https://github.com/pydantic/pydantic-ai|GitHub Repository]]
  * [[https://ai.pydantic.dev|Documentation]]
  * [[https://logfire.pydantic.dev|Pydantic Logfire]]

===== See Also =====

  * [[langchain]] — General-purpose LLM framework
  * [[agno]] — High-performance agent runtime
  * [[haystack]] — Pipeline-based AI orchestration
  * [[google_adk]] — Google's agent development kit (uses Pydantic validation)