Overall UXE Score
8.56
Assessed by the editors · +0.6 vs season average

- Recognition
- Official Selection
- Recognized
- 28 Sept 2026
- How it got here
- Chosen by the editors
- Studio
- Pydantic AI
- Assessed as
- AI & Emerging Experience
- Platforms
- Web
01/Overview
An agent framework whose documentation returns one automated violation across 436 interactive controls.
One violation — a single contrast failure — across 436 interactive controls. That is the highest control count of any page in this round and the lowest fault rate against it. Thirteen landmarks, a skip link, two prefers-reduced-motion rules.
1.55 seconds to largest paint on 702 KB over fifty-two requests, with 14,037 characters of documentation. The layout shift is 0.129, marginally over budget, which is the one soft number here.
The framework's own claim is that agent behaviour should be typed and testable, and the documentation leads with the failure modes rather than a demo. For this category that matters: a page that shows you what goes wrong is disclosing something, and a page that only shows the happy path is not.
Patterns observed
- No consent dialog on first visit
- Client
- In-house product
- Audience
- Python developers building agents they have to debug
- Industry
- AI Products
- Platforms
- Web
- Status
- Live
- Markets
- Worldwide
- Languages
- 1 — English
02/Honors
Recognized by the editors.
Recognition
Official Selection
A reference in its field — consistently excellent across all seven criteria.
Conferred by the editors on 28 Sept 2026 · minimum UXE 8.0
Awards
None. The quarterly awards are decided among projects their makers entered, and nobody entered this one — the editors chose it, assessed it and conferred the recognition above. If Pydantic AI’s team would like it entered and considered for a season, entering is free.
03/Editors’ verdict
In the editors’ words.
“One finding across 436 controls — the best fault rate in the round — and documentation that leads with the failure modes.”
The UXE editors
Selected and assessed by the editors
04/Score breakdown
How the editors scored it.
- Overall UXE Score+0.6 vs Q3 2026 average
- 8.56
- Strongest criterionPerformance
- 8.9
- Assessed bytheir own scores, signed
- The editors
- Measured28 Sept 2026
- On the live product
- 8.6+0.4
Usability
Can people complete what they came to do — efficiently, confidently and without error?
Editors’ noteThe documentation opens with failure modes rather than a demo, which is disclosure rather than decoration.
- 8.8+0.9
Information Architecture
Is content organized, labeled and navigable in the way people actually think?
- 8.6+0.3
UI & Consistency
Is the visual language coherent and purposeful across every screen and state?
- 8.0+0.2
Interaction & Feedback
Does every action respond clearly, promptly and meaningfully?
- 8.9+1.4
Performance
Does it feel fast — perceived and measured — on real devices and networks?
Editors’ note1.55 s to largest paint on 702 KB; layout shift of 0.129, marginally over budget.
- 8.4+0.6
Accessibility
Can everyone use it, including people relying on assistive technology?
Editors’ noteOne serious violation across 436 interactive controls, the highest control count assessed. Thirteen landmarks, a skip link, two reduced-motion rules.
- 8.6+0.5
Learnability
How quickly does a first-time user become a confident one?
Scores are the editors’ own — this product was selected rather than entered, and the verdict is signed. Ticks mark the Q3 2026 average across 161 recognized projects.
05/Screens
2 screensThe product, screen by screen.


06/Accessibility & performance
Built for everyone, on any device.
Accessibility
Assessed here
Audited by the UXE editors (axe-core 4.10 over the loaded page, plus manual checks), September 2026
8.4
Measured
- Pinch zoom is not blocked
- No horizontal scrolling at the assessed viewport
- 13 landmark regions, and a page language declared as en
- A skip link was found
- 2 prefers-reduced-motion rules in the stylesheet
Known gaps
- 1 automated violations in one pass — 1 serious
- Elements must meet minimum color contrast ratio thresholds (1 element, serious)
Performance
3 of 4 within budget
Chrome headless at 1440×900, throttled to Slow 4G (1.6 Mbps down, 150 ms round trip) with 4× CPU slowdown. Three cold loads, each in a fresh browser context with the HTTP cache disabled; the median is reported. 28 September 2026.
8.9
Measured
| Metric | Measured | Budget |
|---|---|---|
| Largest Contentful Paint | 1.55 s (within budget) | ≤ 2.5 s |
| First Contentful Paint | 1.41 s (within budget) | ≤ 1.8 s |
| Cumulative Layout Shift | 0.129 (over budget) | ≤ 0.10 |
| Page weight | 702 KB (within budget) | ≤ 1 MB |
Within budget Over budget
07/Credits
The people who made it.
Studio
Pydantic AI
An agent framework from the team behind the Pydantic data-validation library, developed in public.
- Recognized
- 1
Team
Individual credits are published when the team submits them. This profile was compiled by the editors, so it names the company and nobody else.
Check it yourself
- Client
- In-house product
- Industry
- —
Built with
→/More from the showcase
Whole seasonMore from the showcase.

8.87
UXE

8.83
UXE

8.76
UXE