The Fix
pip install pydantic==2.11.0
Based on closed pydantic/pydantic issue #11574 · PR/commit linked
@@ -1195,6 +1195,13 @@ def __repr__(self) -> str:
def __repr_args__(self) -> _repr.ReprArgs:
+ # Eagerly create the repr of computed fields, as this may trigger access of cached properties and as such
+ # modify the instance's `__dict__`. If we don't do it now, it could happen when iterating over the `__dict__`
+ # below if the instance happens to be referenced in a field, and would modify the `__dict__` size *during* iteration.
repro.py
from __future__ import annotations
from functools import cached_property
from pydantic import BaseModel, computed_field
class Model(BaseModel):
parent: Model | None = None
children: list[Model] = []
@computed_field
@cached_property
def expensive(self) -> bool:
return True
foo = Model()
bar = Model()
foo.children.append(bar)
bar.parent = foo
print(foo)
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==2.11.0\nWhen NOT to use: This fix is not suitable if cached properties should not be eagerly evaluated.\n\nOption C — Workaround\nis to set `@computed_field(repr=True)`. Would this work for you, or do you still want the cached property to be included in the repr? (And thus computed).\nWhen NOT to use: This fix is not suitable if cached properties should not be eagerly evaluated.\n\n
Why This Fix Works in Production
- Trigger: print(foo)
- Mechanism: The cached_property causes a RuntimeError due to dictionary size change during iteration
- Why the fix works: Fixes a RuntimeError when computing model string representation involving cached properties and self-referenced models. (first fixed release: 2.11.0).
Why This Breaks in Prod
- Shows up under Python 3.10 in real deployments (not just unit tests).
- The cached_property causes a RuntimeError due to dictionary size change during iteration
- Surfaces as: Traceback (most recent call last):
Proof / Evidence
- GitHub issue: #11574
- Fix PR: https://github.com/pydantic/pydantic/pull/11579
- First fixed release: 2.11.0
- Reproduced locally: No (not executed)
- Last verified: 2026-02-09
- Confidence: 0.95
- Did this fix it?: Yes (upstream fix exists)
- Own content ratio: 0.41
Discussion
High-signal excerpts from the issue thread (symptoms, repros, edge-cases).
“An alternative workaround is to set @computed_field(repr=True). Would this work for you, or do you still want the cached property to be included in the…”
“Thanks for the quick reply”
“Thanks for the bug report, I went with straightforward fix which should land in 2.11.”
Failure Signature (Search String)
- print(foo)
Error Message
Stack trace
error.txt
Error Message
-------------
Traceback (most recent call last):
File "/home/username/script.py", line 24, in <module>
print(foo)
File "/home/username/miniforge3/envs/pydantic_issue/lib/python3.10/site-packages/pydantic/main.py", line 1127, in __str__
return self.__repr_str__(' ')
File "/home/username/miniforge3/envs/pydantic_issue/lib/python3.10/site-packages/pydantic/_internal/_repr.py", line 61, in __repr_str__
return join_str.join(repr(v) if a is None else f'{a}={v!r}' for a, v in self.__repr_args__())
File "/home/username/miniforge3/envs/pydantic_issue/lib/python3.10/site-packages/pydantic/_internal/_repr.py", line 61, in <genexpr>
return join_str.join(repr(v) if a is None else f'{a}={v!r}' for a, v in self.__repr_args__())
File "/home/username/miniforge3/envs/pydantic_issue/lib/python3.10/site-packages/pydantic/main.py", line 1099, in __repr_args__
for k, v in self.__dict__.items():
RuntimeError: dictionary changed size during iteration
Minimal Reproduction
repro.py
from __future__ import annotations
from functools import cached_property
from pydantic import BaseModel, computed_field
class Model(BaseModel):
parent: Model | None = None
children: list[Model] = []
@computed_field
@cached_property
def expensive(self) -> bool:
return True
foo = Model()
bar = Model()
foo.children.append(bar)
bar.parent = foo
print(foo)
Environment
- Python: 3.10
- Pydantic: 2
What Broke
Attempting to display a model with cached properties results in a RuntimeError.
Why It Broke
The cached_property causes a RuntimeError due to dictionary size change during iteration
Fix Options (Details)
Option A — Upgrade to fixed release Safe default (recommended)
pip install pydantic==2.11.0
When NOT to use: This fix is not suitable if cached properties should not be eagerly evaluated.
Use when you can deploy the upstream fix. It is usually lower-risk than long-lived workarounds.
Option C — Workaround Temporary workaround
is to set `@computed_field(repr=True)`. Would this work for you, or do you still want the cached property to be included in the repr? (And thus computed).
When NOT to use: This fix is not suitable if cached properties should not be eagerly evaluated.
Use only if you cannot change versions today. Treat this as a stopgap and remove once upgraded.
Fix reference: https://github.com/pydantic/pydantic/pull/11579
First fixed release: 2.11.0
Last verified: 2026-02-09. Validate in your environment.
When NOT to Use This Fix
- This fix is not suitable if cached properties should not be eagerly evaluated.
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 |
|---|---|
| 2.11.0 | Fixed |
Related Issues
No related fixes found.
Sources
We don’t republish the full GitHub discussion text. Use the links above for context.