Pydantic
@pydantic
The Pydantic Stack: Pydantic Validation, Pydantic AI, Pydantic Logfire, Pydantic Evals, and Pydantic AI Gateway
172 Following 20.3K Followers
Pydantic AI agents now run on Jev, the classifier from @typesafeai.
Jev doesn't write text, it answers typed questions. So the output_type you already wrote is the question, and the answer comes back as your model, one confidence per field.
Show more
Our founding engineer @Marcelotryle is at AGNTCon + MCPCon Europe in Amsterdam.
Find him if you want to talk MCP, agents, or observability.
He's on stage tomorrow, Sept 18 at 10:20am: Self-Healing Agents Need Observability.
#AGNTCon# #MCPCon#
Show more
Working on a Pydantic AI integration now. Watch this space.
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
Show more
Pydantic AI version 1.107.6 is out! 🎉
Pydantic AI version 2.44.0 is out! 🎉
If you trace LLM calls in one tool and your app in another, agent runs that don't throw errors are hard to debug.
On Sep 30, we'll show how Pydantic Logfire puts the model call and the rest of the request in one OTel trace. Save your spot:
Show more
Picking an image model takes trying a few. That used to cost a new SDK and a new response shape each time to compare outputs.
images = ImageGenerator('openai:gpt-image-2')
Swap the prefix for google:, google-cloud: or xai: and nothing else moves. New from @dasfacc on Pydantic AI:
Show more
Jiter perf vs. the de facto JSON library in Rust.
Significant `jiter` release, in particular this includes SIMD support for x86, which should give up to 7x performance improvement for parsing long strings.
There are also significant improvements for tape reuse, arm64 string SIMD, float parsing and more.
jiter is the most used 3rd party JSON library in Python as a dependency of @pydantic validation.
it's also used in Pydantic Monty, and in Pydantic Logfire via datafusion-functions-json.
Short story: apply strict constraints, provide clear benchmarks for the thing you want to improve, use @codspeedhq, let fable and astra do their thing.
Show more
Another weekend doing open source.
Thanks @FastAPI for sponsoring my work, and @pydantic for giving me a job.
Without you two, I wouldn’t be able to pay my bills, and dedicate time to open source.
Thanks @github and @linuxfoundation for the GitHub Copilot subscription, and @OpenAI and @AnthropicAI for the subscriptions as well.
Without your tokens, it would be hard to keep up with the world.
Show more
Pydantic AI Harness v0.31.0 is out! 🎉
Pydantic AI version 2.43.0 is out! 🎉
> Full Monty commercial service, is ready for production now. LMK if you want a snapshotable, durable, forkable sandbox with <1ms launch times.
We did, and we got it. It's amazing. :)
I go to New York to sell @pydantic Logfire. Easy - it's the best.
Pydantic team wonder how long it'll take for enterprise to even think about Pydantic Monty (it's Samuel's crazy side project, brand new, is it even useful?), so I ask: ALL the financial institutions we speak to have teams (whole fucking TEAMS!) building Monty as a service for internal use.
Multiple massive financial institutions are spending millions to enable them to adopt one sandboxing technology. We thought we were too early to build a product around Monty, we're late.
Full Monty commercial service, is ready for production now. LMK if you want a snapshotable, durable, forkable sandbox with <1ms launch times.
Show more
AWS Lambda durable functions now integrates with @pydantic AI harness
Now, you can use gh aw to let your custom @pydantic
AI agent maintain your repo.
Have you tried the GitHub agentic workflows engine?
Build your own coding harness with Pydantic AI harness and Codex subagents.
Pydantic AI Harness now ships AWS LambdaDurability.
Durable execution on your Lambda functions with Pydantic AI so a retry resumes from a checkpoint.
Pydantic AI Harness v0.30.0 is out! 🎉
Pydantic AI version 2.42.0 is out! 🎉
I'm in NYC, in the country of big portions, and now I get the smallest "panini" I've ever seen.
Sad and confused.
Hand for scale.
