pydantic/pydantic-ai
A Python agent framework organized around typed dependencies, structured outputs, tools, model providers, durable execution, evaluation, and observability.
Python developers building typed agent applications.
Python library / typed agent and tool APIs / evaluation and observability integrations
Types can make model output and tool contracts explicit
The framework presents agent dependencies, tool inputs, and structured results through Python types and validation boundaries.
Reporting is still in progress.
This scope profile names the project's place in the field and one useful mechanism. It does not claim that every runtime boundary has been inspected.
Learn Through This Project
7 lessonsLatest Reporting
8 articlesOctober 14, 2026 — Google’s Argon can continue a response beyond one request’s output limit. Pydantic AI’s released integration shows what that means for an agent: one answer, several network calls, and two different ways of counting tokens. pydantic/pydantic-ai
Who Supplies the Agent? OpenAI Hosts It, Pydantic Assembles ItSeptember 28, 2026 — Pydantic’s weekend changes tie a framework, a coding harness and a terminal client to one release process. Set beside OpenAI’s managed Agents API, they reveal a choice that model rankings miss: which parts of the agent your team still has to run. openai/codex, pydantic/pydantic-ai, openclaw/openclaw
API Discovery: Let the Agent Pick a Record You Already HaveSeptember 20, 2026 — Pydantic AI’s new Choices helper builds a menu from data available at run time. The model selects a validated key; the application gets back the record that key represents. pydantic/pydantic-ai
API Discovery: Let the Speaker Finish Before You CloseSeptember 19, 2026 — Pydantic AI adds a session-level wait between a generated reply and closing the voice connection. The audio consumer still has to tell the truth about playback. pydantic/pydantic-ai
API Discovery: Tell a Running Pydantic Agent What Just ChangedSeptember 16, 2026 — Pydantic AI’s enqueue API lets a worker add context while an agent is running. A September 15 repair makes the closing boundary explicit: the message enters the queue, or the caller gets an error. pydantic/pydantic-ai
API Discovery: Let a Pydantic AI Agent Use Your Codex SubscriptionSeptember 8, 2026 — Pydantic AI’s new Codex provider turns an existing subscription into a typed-agent option—with its authentication and streaming limits in plain view. pydantic/pydantic-ai
API Discovery: Let an Agent Notice When Its Reasoning Was DroppedSeptember 3, 2026 — Pydantic AI now carries an Anthropic thinking-block fallback through later turns—but the useful operator move is to expose the loss, then fix the unstable prompt prefix that caused it. pydantic/pydantic-ai
API Discovery: Tool Calls Need a Home in the StreamAugust 30, 2026 — Pydantic AI’s public AG-UI adapter now gives a tool-only response an announced assistant-message parent—useful when a frontend rebuilds a conversation from streamed events. pydantic/pydantic-ai
Primary Sources
Open Questions
- Which guarantees are truly enforced by types, and which still depend on application policy and runtime authority?
Evidence state describes what this Atlas profile can support today. It is not a product score and it is not evidence that uninspected capabilities are absent.