Skip to content

__slots__

In one line

__slots__ declares fixed attributes, eliminating __dict__ and saving ~40-50% memory per instance.

In simple words

By default, Python objects store attributes in a per-instance __dict__ (a hash table). This is flexible but memory-heavy. For classes with millions of instances, this adds up.

Defining __slots__ = ['x', 'y', 'z'] tells Python to use a fixed-size array instead of a dict. Benefits:

  • Memory savings — ~40-50% less per instance (no dict overhead)
  • Faster attribute access — array indexing vs hash lookup
  • Accidental attribute prevention — assigning an unlisted attribute raises AttributeError

Trade-offs: you lose __dict__, can't easily add new attributes, and some features (multiple inheritance with slots) are tricky. Use when you have many instances of the same class.

Example

python
import sys

class PointDict:
    def __init__(self, x, y):
        self.x = x
        self.y = y

class PointSlots:
    __slots__ = ["x", "y"]
    def __init__(self, x, y):
        self.x = x
        self.y = y

p1 = PointDict(1, 2)
p2 = PointSlots(1, 2)

print(sys.getsizeof(p1.__dict__))  # ~104 bytes (dict)
print(sys.getsizeof(p2))          # Smaller (no dict)
# p2.z = 3  # AttributeError! Not in slots

__slots__ replaces __dict__ with a fixed array — less memory, faster access, but you can't add undeclared attributes.

Common confusions

  • Confused with: frozen dataclass

    The difference: __slots__ is about memory optimization — it restricts attributes but they're still mutable. frozen=True on a dataclass is about immutability — attributes can't be changed but memory isn't necessarily optimized. You can combine both: @dataclass(frozen=True, slots=True).

Related terms

Related Resources

Related Lessons

Decode error messages

Plain-English explanations of common errors — what they mean, why they happen, and how to fix them.

View errors

← Back to glossary