__slots__

In Python, __slots__ is a class attribute that names the only instance attributes a class’s objects can have, so Python stores them in a fixed, compact layout instead of a per-instance dictionary. You assign it a tuple or other iterable of attribute names in the class body.

By default, every instance keeps its attributes in its own __dict__, a dictionary that lets you add new names at any time. Declaring __slots__ removes that dictionary. Python instead creates a descriptor on the class for each listed name, and every instance reserves room for exactly those values:

A regular instance's __dict__ stores any name, even a typo. A slotted instance has fixed slots and raises AttributeError.
A Growable __dict__ Versus Fixed Slots

That trade has two visible effects. Instances use noticeably less memory, which adds up when you create millions of small objects, and attribute lookups can get faster too. Assigning to a name that isn’t listed raises an AttributeError, so misspelled attribute names fail loudly.

A few rules come with the feature. Slotted instances don’t support weak references unless you add "__weakref__" to the names, and a class attribute can’t supply a default value for a slot of the same name.

A subclass gets a __dict__ back unless it also defines __slots__, listing only the names it adds, if any. Since Python 3.10, the dataclasses module can generate the declaration for you with @dataclass(slots=True).

Example

Say you’re loading millions of GPS readings into memory, and you want each one to stay small and reject misspelled field names:

Python 
>>> class Reading:
...     __slots__ = ("latitude", "longitude")
...     def __init__(self, latitude, longitude):
...         self.latitude = latitude
...         self.longitude = longitude
...

>>> reading = Reading(52.52, 13.40)
>>> reading.latitude
52.52

>>> reading.lattitude = 48.85
Traceback (most recent call last):
  ...
AttributeError: 'Reading' object has no attribute 'lattitude' and no __dict__...

>>> hasattr(reading, "__dict__")
False

Because Reading declares __slots__, each instance holds .latitude and .longitude in fixed slots and carries no __dict__. On a regular class, the misspelled .lattitude would create a third attribute without any error. Here, Python rejects it, and the missing dictionary is where the memory savings come from.

Python Classes: The Power of Object-Oriented Programming

Tutorial

Python Classes: The Power of Object-Oriented Programming

Learn how to define and use Python classes to implement object-oriented programming. Dive into attributes, methods, inheritance, and more.

intermediate best-practices python

For additional information on related topics, take a look at the following resources:

Have a question about this? Mentor AI can show you examples, compare related terms, and point you to tutorials.

  • Ask Mentor AI
  • Pick a question and Mentor AI takes it from here.
By Martin Breuss • Updated Oct. 31, 2026
Python Glossary Share Feedback