@plutaniano,

Huh, this is definitely a bug. Thanks for reporting this! Likely an issue due to undesired rounding. A PR with a fix is more than welcome!

Hi, I've been investigating this today, seems that using decimal.Decimal instead of float makes the trick.

from typing import Annotated
from pydantic import TypeAdapter, Field
from decimal import Decimal

for i in range(-10, 10):
    x = i / 100
    try:
        TypeAdapter(Annotated[Decimal, Field(multiple_of=0.01)]).validate_python(x)
        print(f"  OK: {x}")
    except:
        print(f"FAIL: {x}")

Results in:

  OK: -0.1
  OK: -0.09
  OK: -0.08
  OK: -0.07
  OK: -0.06
  OK: -0.05
  OK: -0.04
  OK: -0.03
  OK: -0.02
  OK: -0.01
  OK: 0.0
  OK: 0.01
  OK: 0.02
  OK: 0.03
  OK: 0.04
  OK: 0.05
  OK: 0.06
  OK: 0.07
  OK: 0.08
  OK: 0.09

This should be fixed at pydantic-core though.

@sydney-runkle
Hi!
It looks like the validation in pydantic-core needs some modification. If this issue is still unresolved, could you please assign it to me? I have a good idea of where the changes are needed, so I believe I can submit a PR relatively quickly, including additional test cases.

Hi @K-dash,
I was looking at the same issue, but you made it quicker:).

Just wanted to point out that IMO - to maintain coherence with the existing setting - you might want to add the same test parametrizations of pydantic-core/tests/validators/test_float.py::test_float_multiple_of to pydantic-core/tests/validators/test_decimal.py::test_decimal_multiple_of and make sure tests pass.

Hi @AlessandroMiola
Thank you for your suggestion:).
I agree with your point, so I have added the same test cases to test_decimal_multiple_of as well.