Skip to main content

Caching Small Integers in Python

Python implements small integer caching as an optimization technique to improve performance and memory usage. Here are the key points about small integer caching in Python:

Range of Cached Integers​

Python caches integer values in the range of -5 to 256 (inclusive). This means that for integers within this range, Python will reuse existing objects rather than creating new ones each time.

How It Works​

When you create an integer within the cached range, Python returns a reference to an existing pre-allocated object instead of creating a new one. This is done automatically by the Python interpreter during startup.

Performance Benefits​

Caching small integers provides performance benefits because:

  1. These numbers are used very frequently in programs.
  2. It saves time and memory by avoiding repeated object creation for common integer values.

Behavior Differences​

The caching behavior can lead to some unexpected results when comparing integers:

a = 256
b = 256
print(a is b)  # True

c = 257
d = 257
print(c is d)  # False

In this example, a and b refer to the same cached object, while c and d are separate objects.

Important Considerations​

  1. This behavior is specific to the CPython implementation and may differ in other Python implementations.

  2. The caching is an implementation detail and should not be relied upon in your code.

  3. Always use the == operator for comparing integer values, not the is operator:

# Correct way to compare integers
a = 257
b = 257
print(a == b)  # True