Pydantic Documentation should include examples (Fix)

The Fix

pip install pydantic==2.11.8

Based on closed pydantic/pydantic issue #12115 · PR/commit linked

Production note: This usually shows up under retries/timeouts. Treat it as a side-effect risk until you can verify behavior with a canary + real traffic.

Jump to Verify Open PR/Commit

@@ -326,18 +326,15 @@ def collect_model_fields(  # noqa: C901
                 # extend this to any descriptor in the future (by simply checking for
                 # `hasattr(assigned_value.default, '__get__')`).
-                assigned_value.default = assigned_value.default.__get__(None, cls)
-
-            # The `from_annotated_attribute()` call below mutates the assigned `Field()`, so make a copy:

repro.py

from pydantic import BaseModel, EmailStr

class MyModel(BaseModel):
    work_email: EmailStr
    home_email: Optional[EmailStr] = None
    phone_number: str

class MyModelFixed(BaseModel):
    work_email: str
    home_email: Optional[str] = None
    phone_number: str

verify

Re-run the minimal reproduction on your broken version, then apply the fix and re-run.

fix.md

Option A — Upgrade to fixed release\npip install pydantic==2.11.8\nWhen NOT to use: This fix should not be applied if the existing model relies on deprecated behavior.\n\n

Why This Fix Works in Production

Trigger: A little context: I was trying to write some code that creates a "stripped down version" of a schema to be used with Gemini
Mechanism: Refactor of `FieldInfo` creation logic led to improved metadata handling
Why the fix works: Refactor `FieldInfo` creation implementation to improve merging logic and avoid buggy metadata handling. (first fixed release: 2.11.8).

Why This Breaks in Prod

Refactor of `FieldInfo` creation logic led to improved metadata handling
Production symptom (often without a traceback): A little context: I was trying to write some code that creates a "stripped down version" of a schema to be used with Gemini

Proof / Evidence

GitHub issue: #12115
Fix PR: https://github.com/pydantic/pydantic/pull/11898
First fixed release: 2.11.8
Reproduced locally: No (not executed)
Last verified: 2026-02-09
Confidence: 0.95
Did this fix it?: Yes (upstream fix exists)
Own content ratio: 0.56

Discussion

High-signal excerpts from the issue thread (symptoms, repros, edge-cases).

Jump to Sources Open on GitHub

“Can you provide more information on your use case? I'm having trouble following why you are using the FieldInfo class directly”

@Viicos · 2025-08-02 · source

“The schema with the modifications described above works in Gemini. I just didn't want to manually rewrite (duplicate) a whole set of schema classes, but…”

@DavidNemeskey · 2025-08-12 · source

“Basically, I would like to replace each occurrence of EmailStr with str”

@DavidNemeskey · 2025-08-02 · source

“I'm going to close this issue”

@Viicos · 2025-08-12 · source

Failure Signature (Search String)

A little context: I was trying to write some code that creates a "stripped down version" of a schema to be used with Gemini
By _schema_, what are you referring to exactly? What kind of input are you providing to Gemini?

Copy-friendly signature

signature.txt

Failure Signature
-----------------
A little context: I was trying to write some code that creates a "stripped down version" of a schema to be used with Gemini
By _schema_, what are you referring to exactly? What kind of input are you providing to Gemini?

Error Message

Signature-only (no traceback captured)

error.txt

Error Message
-------------
A little context: I was trying to write some code that creates a "stripped down version" of a schema to be used with Gemini
By _schema_, what are you referring to exactly? What kind of input are you providing to Gemini?

Minimal Reproduction

repro.py

from pydantic import BaseModel, EmailStr

class MyModel(BaseModel):
    work_email: EmailStr
    home_email: Optional[EmailStr] = None
    phone_number: str

class MyModelFixed(BaseModel):
    work_email: str
    home_email: Optional[str] = None
    phone_number: str

What Broke

Users faced difficulties in creating models due to unclear documentation and buggy metadata handling.

Why It Broke

Refactor of `FieldInfo` creation logic led to improved metadata handling

Fix Options (Details)

Option A — Upgrade to fixed release Safe default (recommended)

pip install pydantic==2.11.8

When NOT to use: This fix should not be applied if the existing model relies on deprecated behavior.

Use when you can deploy the upstream fix. It is usually lower-risk than long-lived workarounds.

Fix reference: https://github.com/pydantic/pydantic/pull/11898

First fixed release: 2.11.8

Last verified: 2026-02-09. Validate in your environment.

When NOT to Use This Fix

This fix should not be applied if the existing model relies on deprecated behavior.

Verify Fix

verify

Re-run the minimal reproduction on your broken version, then apply the fix and re-run.

Did This Fix Work in Your Case?

Quick signal helps us prioritize which fixes to verify and improve.

Prevention

Add a CI check that diffs key outputs after upgrades (OpenAPI schema snapshots, JSON payload shapes, CLI output).
Upgrade behind a canary and run integration tests against the canary before 100% rollout.

Version Compatibility Table

Version	Status
2.11.8	Fixed

Related Issues

No related fixes found.

Cluster: pydantic:openapi-schema Pydantic hub Pydantic best practices All hubs All clusters

Related clusters: Configuration error Data consistency Serialization

Sources

We don’t republish the full GitHub discussion text. Use the links above for context.