Benchmark list append vs comprehension in Python

This micro-benchmark compares the speed of building a list with a for loop and append versus a list comprehension, using the timeit module to get precise timings.

Easy Python 3.9+ Aug 9, 2026 Functions & basics 13 views 0 copies

timeit benchmark performance list-comprehension micro-benchmark

Python code

24 lines

Python 3.9+

import timeit

# Build a list of the first 1,000,000 integers using append in a loop
def append_loop(n=1_000_000):
    result = []
    for i in range(n):
        result.append(i)
    return result

# Build the same list using a list comprehension
def comprehension(n=1_000_000):
    return [i for i in range(n)]

if __name__ == "__main__":
    n = 1_000_000
    repeat = 5

    # Time both approaches (best of 3 runs, 1 repetition each)
    append_time = min(timeit.repeat(lambda: append_loop(n), number=1, repeat=repeat))
    comp_time = min(timeit.repeat(lambda: comprehension(n), number=1, repeat=repeat))

    print(f"append loop:      {append_time:.4f} seconds")
    print(f"list comprehension: {comp_time:.4f} seconds")
    print(f"comprehension is {append_time / comp_time:.2f}x faster")

Output

stdout

append loop:      0.0950 seconds
list comprehension: 0.0525 seconds
comprehension is 1.81x faster

How it works

The timeit.repeat function runs each function multiple times and returns the timings; taking the minimum reduces noise from system variations. List comprehensions execute the loop in C, avoiding repeated Python method calls like list.append, which makes them faster for creating lists from simple expressions. The speed difference comes from how the bytecode is executed: the comprehension's loop is optimized in the interpreter, while a for loop with an append method has more overhead.

For this one-million-element list, the comprehension consistently wins, typically being about 1.5 to 2 times faster. The exact numbers vary by machine, but the relative performance gap remains.

Common mistakes

Using `timeit.timeit` with `number=1` only once, which is noisy; use `repeat` and take the minimum.
Forgetting that `timeit` measures wall-clock time, so other processes can skew results.
Assuming the comprehension is always faster for complex expressions; it's fastest when the body is a simple expression.

Variations

Use `timeit.timeit` with `number=100` and average to measure more stable timings.
Compare other collection types, like building a dictionary or set with comprehension vs loop.

Real-world use cases

Optimizing a hot path that constructs large lists repeatedly, like building in-memory data for analytics.
Deciding between a loop and comprehension when writing performance-sensitive code in a data pipeline.
Teaching or documenting relative speed of Python constructs during code reviews or performance audits.

Benchmark list append vs comprehension in Python

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Functions & basics

Tutorials

Quizzes

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Functions & basics

Keep learning

Tutorials

Quizzes