Pydantic (@pydantic) / X

Pydantic

2,578 posts

Pydantic

@pydantic

The Pydantic Stack: Pydantic Validation, Pydantic AI, Pydantic Logfire, Pydantic Evals, and Pydantic AI Gateway

London

Joined August 2022

Pinned
Pydantic
@pydantic
Apr 1
AI observability can cost more than it should. We ran the numbers: same scale, same workload, four tools. At moderate production load, Logfire is up to 40x less expensive than the alternatives. Spans-based pricing. No proprietary units. AI native and full-stack observability in
22K
Pydantic
@pydantic
2h
AI features can quietly get worse, and a customer often notices before you do. No errors, no alerts. Latency and token spend look fine because none measures whether the answer was good. If you're building production AI, join us online on Thu, Jun 25 to learn why AI
213
Pydantic reposted
MX3 Dev
@Mx3Dev
Jun 10
@pydantic shipped Pydantic AI v1.107.0 with key fixes and new model support Security fix patches file read risk in UI adapters. Only affects apps passing untrusted client history with vulnerable storage refs. Most users safe.
381
Pydantic reposted
Daniel Himmelstein
@dhimmel
Jun 10
The end of having to use jinja for templating is in sight!? Just starting to play around with @pydantic handlebars github.com/pydantic/pydan… Much respect to #jinja2, but it's always been a tightrope walk to get the logic and whitespace to work as intended.
GitHub - pydantic/pydantic-handlebars
From github.com
547
Pydantic
@pydantic
Jun 10
Pydantic AI version 2.0.0b7 is out! 🎉
Release v2.0.0b7 (2026-06-10) · pydantic/pydantic-ai
From github.com
464
Pydantic
@pydantic
Jun 10
Pydantic AI version 1.107.0 is out! 🎉
Release v1.107.0 (2026-06-10) · pydantic/pydantic-ai
From github.com
658
Pydantic reposted
How To AI
@HowToAI_
Jun 9
Docker for AI Agents is officially over 🤯 Pydantic open-sourced a new way to run LLM-generated code that: - does not need Docker. - does not spin up containers. - does not call any cloud sandbox. - does not cost a cent to run. It's called Monty. Instead of spinning up a
42K
Pydantic reposted
Luísa Lima
@__luisalima__
Jun 9
Ok this is poetry. Thanks @pydantic team
How To AI
@HowToAI_
Jun 9
Docker for AI Agents is officially over 🤯 Pydantic open-sourced a new way to run LLM-generated code that: - does not need Docker. - does not spin up containers. - does not call any cloud sandbox. - does not cost a cent to run. It's called Monty. Instead of spinning up a
3.1K
Pydantic
@pydantic
Jun 9
Your AI tools spend just outgrew a senior engineer's salary and nobody owns it. Our latest blogpost makes the case that "cost control" is really an observability problem in a CFO hat. The fix is a trace, not a cap. 🔗
AI spend is the new headcount: why cost control is an observability problem
From pydantic.dev
925
Pydantic reposted
ClickHouse
@ClickHouseDB
Jun 8
PyPI May 2026: 163.8 billion downloads (+9.2% MoM). The AI agent race is now measurable in download counts: ⚡ sglang: 304M (+744% in 3 months), @lsyincs @sgl_project - the sleeper hit 🔀 litellm: 513M (+425% in 3 months), @krrish_dh @ishaan_jaff @LiteLLM 📐 pydantic-ai-slim:
26K
Pydantic
@pydantic
Jun 8
Our founder @samuelcolvin sat down with @temporalio folks to chat about the importance of durable execution and type-safety for production AI agents.
Why Most Agentic AI Projects Never Leave the Pilot Phase – Temporal Ep 2
From techstrong.tv
1.5K
Pydantic
@pydantic
Jun 8
eval-type-backport is a library that makes modern type hint syntax like `int | str` "just work" in older Python versions. In today's blogpost, ;earn how the library works. Bonus: the author explains why metaprogramming is a skill worth learning in the AI era and how it helped
1K
Pydantic
@pydantic
Jun 5
Replying to @pydantic @samuelcolvin and @pydatalondon
Make sure to stick around for @summerscope's talk on Sunday: "The human in the loop is tired". Laura will look into what actually feels like to work with code in the age of AI. Good, bad, and whimsical. Based her most popular blogpost:
The Human-in-the-Loop is Tired
From pydantic.dev
863