line_profiler
line_profiler is a line-by-line performance profiler for Python that measures execution time for each line within selected functions and reports detailed timing statistics.
Installation and Setup
Install from PyPI:
If you want the IPython/Jupyter magics too, install the extra:
$ python -m pip install "line_profiler[ipython]"
Key Features
- Provides line-by-line timing reports with hit counts, total time, per-hit time, and per-line percentage breakdown.
- Ships the
kernprofrunner to execute your program and write profiling results to a.lproffile. - Offers a CLI viewer to inspect
.lprofoutput and control display options like output units. - Integrates with IPython and Jupyter via
%load_ext line_profilerand%lprunfor ad hoc profiling in notebooks. - Provides a Python API for profiling function calls directly in code and printing stats with a chosen
output_unit.
Usage
Mark functions to profile by adding the @profile decorator:
script.py
from line_profiler import profile
@profile
def work(n=10000):
total = 0
for i in range(n):
total += i * i
return total
if __name__ == "__main__":
work()
The quickest way to run the profiler agains the script is to use the LINE_PROFILE environment variable:
$ LINE_PROFILE=1 python script.py
Timer unit: 1e-09 s
0.00 seconds - .../script.py:4 - work
Wrote profile results to profile_output.txt
Wrote profile results to profile_output_2025-12-15T124644.txt
Wrote profile results to profile_output.lprof
To view details run:
python -m line_profiler -rtmz profile_output.lprof
View the results:
python -m line_profiler -rtmz profile_output.lprof
Timer unit: 1e-06 s
Total time: 0.004837 s
File: .../script.py
Function: work at line 4
Line # Hits Time Per Hit % Time Line Contents
==============================================================
4 @profile
5 def work(n=10000):
6 1 1.0 1.0 0.0 total = 0
7 10001 2466.0 0.2 51.0 for i in range(n):
8 10000 2369.0 0.2 49.0 total += i * i
9 1 1.0 1.0 0.0 return total
0.00 seconds - .../script.py:4 - work
Run with kernprof and save the results to a .lprof file, then display them:
$ kernprof -l script.py
Wrote profile results to 'script.py.lprof'
Inspect results with:
python -m line_profiler -rmt script.py.lprof
$ python -m line_profiler script.py.lprof
Timer unit: 1e-06 s
Total time: 0.003005 s
File: script.py
Function: work at line 4
Line # Hits Time Per Hit % Time Line Contents
==============================================================
4 @profile
5 def work(n=10000):
6 1 0.0 0.0 0.0 total = 0
7 10001 1431.0 0.1 47.6 for i in range(n):
8 10000 1573.0 0.2 52.3 total += i * i
9 1 1.0 1.0 0.0 return total
0.00 seconds - script.py:4 - work
Use the programmatic API:
fib.py
from line_profiler import LineProfiler
def fib(n):
a, b = 0, 1
for _ in range(n):
a, b = b, a + b
return a
lp = LineProfiler()
lp.add_function(fib)
lp_wrapper = lp(fib)
lp_wrapper(100000)
lp.print_stats(output_unit=1e-3)
Related Resources
Tutorial
Profiling in Python: How to Find Performance Bottlenecks
In this tutorial, you'll learn how to profile your Python programs using numerous tools available in the standard library, third-party libraries, as well as a powerful tool foreign to Python. Along the way, you'll learn what profiling is and cover a few related concepts.
For additional information on related topics, take a look at the following resources:
By Leodanis Pozo Ramos • Updated Dec. 15, 2025
