Python Developers Survey 2024 Results
This is the eighth annual Python Developers Survey, conducted as a collaborative effort between the Python Software Foundation and JetBrains.
Responses were collected in October and November 2024, with more than 30,000 Python developers and enthusiasts from almost 200 countries and regions taking part to illuminate the current state of the language and its ecosystem.
Check out the Python Developer Survey results from 2023, 2022, 2021, 2020, 2019, 2018, and 2017.
Share:
General Python Usage
Python as main vs secondary language
Main
Secondary
Python usage with other languages100+
All options with less than 2% have been merged into “Other”.
Languages for Data Science and Web100+
45%
46%
SQL
34%
64%
JavaScript
32%
54%
HTML/CSS
31%
35%
Bash / Shell
29%
16%
C/C++
19%
13%
Java
14%
29%
TypeScript
11%
2%
R
10%
8%
C#
Web development refers to people who selected Web development in response to the question “What do you use Python for the most?”
Data science refers to people who selected Data analysis or Machine learning in the same question.
Python usage with other languages100+
40%
43%
JavaScript
37%
35%
HTML/CSS
37%
30%
SQL
32%
25%
Bash / Shell
26%
37%
C/C++
17%
28%
Java
15%
24%
TypeScript
9%
18%
C#
8%
12%
Go
of surveyed Python developers practice collaborative development, down 7 percentage points from last year.
This decline may be due to remote work fatigue, with developers preferring individual workflows, or the return to office environments, where collaboration dynamics shift.
How many years of professional coding experience do you have?
Less than 1 year
1—2 years
3—5 years
6—10 years
11+ years
How long have you been programming in Python?
Less than 1 year
1—2 years
3—5 years
6—10 years
11+ years
Did you know that Python is the most popular language for learning to code?
One in five surveyed respondents has been programming in Python for less than a year, and over two-thirds of computer science learners worldwide reported using Python for both learning and work in the past year.
Check out our Computer Science Learning Curve Survey 2024 Report to explore current trends – from learning formats and tools to motivations, career goals, and common challenges.
of Pythonistas reported contributing to open-source projects last year.
In the past year, how would you describe your contributions to open source?100+
Where do you typically learn about new tools and technologies that are relevant to your Python development?100+
55%
58%
Documentation and APIs
45%
51%
YouTube
44%
41%
Python.org
42%
43%
Stack Overflow
41%
38%
Blogs
28%
22%
Books
19%
27%
AI Tools
14%
13%
Online coding schools and MOOCs
AI is gaining popularity as a method of learning about new tools and technologies in Python. From 2023 to 2024, the proportion of learners who report using AI for this purpose rose from 19% to 27%.
Purposes for Using Python
In this section, we asked questions to find out what people use Python for, what types of development they are involved in, and how they combine their various uses.
For what purposes do you mainly use Python?
Both for work and personal
For personal, educational or side projects
For work
What do you use Python for?100+
Python usage as main and secondary language100+
49%
42%
Data analysis
48%
34%
Web development
42%
33%
Machine learning
33%
22%
Data engineering
28%
23%
Web scraping & parsing
28%
23%
Academic research
26%
25%
DevOps / Systems administration
What do you use Python for the most?
21%
23%
Web development
10%
13%
Machine learning
10%
10%
Data analysis
9%
8%
Academic research
9%
8%
Educational purposes
7%
6%
DevOps / systems administration / writing automation scripts
6%
7%
Data engineering
In this question, we asked respondents to select only one primary activity.
To what extent are you involved in the following activities?
Web development
Data analysis
Machine learning
Data engineering
Academic research
DevOps / systems administration / writing automation scripts
Educational purposes
Software testing / Writing automated tests
Software prototyping
Design / Data visualization
Programming of web parsers / scrapers / crawlers
Desktop development
Network programming
Python Versions
of surveyed Python developers continue to use Python 2.
Python 3 versions
Python installation and upgrade100+
Why haven’t you updated to the latest version?100+
Frameworks and Libraries
Web frameworks100+
Percentages are calculated within each column.
| 2021 | 2022 | 2023 | 2024 | |
|---|---|---|---|---|
| 21% | 25% | 29% | 38% | FastAPI |
| 40% | 39% | 33% | 35% | Django |
| 41% | 39% | 33% | 34% | Flask |
| – | – | 30% | 33% | Requests |
| – | – | 20% | 23% | Asyncio |
| – | – | 18% | 20% | Django REST Framework |
| – | – | 12% | 15% | httpx |
| – | – | 12% | 13% | aiohttp |
| – | – | 8% | 12% | Streamlit |
| – | – | 6% | 8% | Starlette |
| 3% | 4% | 3% | 3% | web2py |
| 4% | 4% | 3% | 2% | Tornado |
| 3% | 3% | 3% | 2% | Bottle |
| 3% | 4% | 3% | 2% | CherryPy |
| 3% | 3% | 3% | 2% | Pyramid |
| 2% | 2% | 2% | 1% | Falcon |
| 1% | 2% | 1% | 1% | Hug |
| – | – | 2% | 1% | Quart |
| – | – | 2% | 1% | Twisted |
| 5% | 5% | 5% | 7% | Other |
| 29% | 27% | 23% | 19% | None |
All options with less than 2% have been merged into “Other”.
Web frameworks100+
41%
56%
FastAPI
37%
39%
Flask
33%
42%
Requests
28%
61%
Django
22%
33%
Asyncio
22%
7%
Streamlit
13%
44%
Django REST Framework
Web frameworks cross-usage100+
Percentages are calculated within each column.
| Asyncio | Django | Django REST Framework | FastAPI | Requests | Starlette | Streamlit | aiohttp | httpx | |
|---|---|---|---|---|---|---|---|---|---|
| – | 26% | 33% | 42% | 45% | 69% | 37% | 81% | 56% | Asyncio |
| 38% | – | 93% | 42% | 41% | 37% | 38% | 39% | 38% | Django |
| 27% | 53% | – | 29% | 28% | 27% | 23% | 28% | 26% | Django REST Framework |
| 68% | 45% | 55% | – | 55% | 92% | 65% | 67% | 69% | FastAPI |
| 43% | 47% | 47% | 45% | 47% | 35% | 51% | 42% | 36% | Flask |
| 62% | 39% | 47% | 48% | – | 67% | 54% | 64% | 56% | Requests |
| 23% | 8% | 11% | 19% | 16% | – | 15% | 24% | 27% | Starlette |
| 19% | 13% | 14% | 21% | 19% | 22% | – | 17% | 17% | Streamlit |
| 45% | 15% | 19% | 23% | 25% | 41% | 19% | – | 35% | aiohttp |
| 35% | 16% | 20% | 27% | 25% | 52% | 21% | 40% | – | httpx |
| 21% | 18% | 18% | 18% | 20% | 22% | 20% | 24% | 27% | Other |
You can find more about the Django landscape in the Django Developers Survey 2023, conducted in partnership with the Django Software Foundation.
Other frameworks and libraries100+
31%
34%
BeautifulSoup
28%
32%
Pillow
22%
30%
Pydantic
22%
26%
OpenCV-Python
17%
21%
Tkinter
12%
13%
PyQT
11%
12%
Scrapy
10%
11%
Pygame
Unit-testing frameworks100+
pytest
unittest
mock
doctest
tox
Hypothesis
nose
Other
None
For which frameworks would you like to have rich support in your editor / IDE?100+
PyCharm
PyCharm provides extended support for backend development, including support for Django, FastAPI, and Flask.
Cloud platforms
Cloud platforms usage100+
Please note that in 2023, the list was expanded with new options.
All options with less than 2% have been merged into “Other”.
How do you run code in the cloud?100+
Within containers
In virtual machines
Serverless
On a platform-as-a-service
Other
None
of surveyed developers use Kubernetes for running code in containers.
Which of the following do you use?100+
Amazon Elastic Kubernetes Service
Google Kubernetes Engine
Azure Kubernetes Service
RedHat OpenShift
Other
How do you develop for the cloud?100+
49%
51%
Locally with virtualenv
38%
44%
In Docker containers
23%
23%
In virtual machines
20%
19%
With local system interpreter
16%
16%
In remote development environments
14%
15%
Using WSL
10%
9%
Directly in the production environment
2%
2%
Other
Data Science
of all surveyed Python developers are involved in data exploration and processing, with pandas and NumPy being the tools mostly used for it.
Tools for data exploration and processing
All options with less than 2% have been merged into “Other”.
Tools for data versioning
An in-house solution
Delta lake
DVC
Pachyderm
Other
None
of surveyed Pythonistas reported that they work on creating dashboards, with Streamlit and Plotly Dash being the top choices for such tasks.
Libraries for creating dashboards100+
Streamlit
Plotly Dash
TensorBoard
Gradio
Panel
Voila
Other
None
BI solutions100+
PowerBI
I'm not sure
Tableau
Looker
Metabase
QlikView
Other
None
All options with less than 2% have been merged into “Other”.
of our respondents train or generate predictions using ML models, which is an increase of six percentage points from last year. Among them, more than two thirds use scikit-learn and PyTorch.
Frameworks for ML model training and prediction100+
67%
68%
SciKit-Learn
60%
66%
PyTorch
48%
49%
TensorFlow
44%
42%
SciPy
30%
30%
Keras
22%
28%
Hugging Face Transformers
22%
23%
XGBoost
All options with less than 2% have been merged into “Other”.
Frameworks for ML model training or prediction cross-usage100+
Percentages are calculated within each column.
| Hugging Face Diffusers | Hugging Face Transformers | Keras | NLTK | PyTorch | PyTorch Lightning | SciKit-Learn | SciPy | TensorFlow | XGBoost | spaCy | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| – | 38% | 18% | 22% | 16% | 25% | 14% | 16% | 17% | 17% | 25% | Hugging Face Diffusers |
| 90% | – | 38% | 53% | 37% | 46% | 33% | 34% | 34% | 42% | 62% | Hugging Face Transformers |
| 47% | 40% | – | 50% | 36% | 37% | 41% | 42% | 52% | 50% | 46% | Keras |
| 36% | 36% | 33% | – | 24% | 28% | 27% | 29% | 27% | 35% | 59% | NLTK |
| 88% | 86% | 78% | 80% | – | 94% | 72% | 77% | 76% | 75% | 82% | PyTorch |
| 31% | 24% | 18% | 21% | 21% | – | 18% | 21% | 16% | 21% | 25% | PyTorch Lightning |
| 74% | 78% | 89% | 90% | 73% | 79% | – | 91% | 80% | 94% | 88% | SciKit-Learn |
| 57% | 50% | 59% | 62% | 49% | 61% | 58% | – | 52% | 62% | 68% | SciPy |
| 69% | 59% | 85% | 68% | 57% | 55% | 59% | 61% | – | 63% | 63% | TensorFlow |
| 33% | 34% | 38% | 42% | 26% | 34% | 34% | 34% | 30% | – | 43% | XGBoost |
| 30% | 31% | 22% | 43% | 18% | 24% | 19% | 23% | 18% | 26% | – | spaCy |
Experiment tracking tools100+
TensorBoard
MLFlow
Weights & Biases
An in-house solution
NeptuneML
CometML
Other
None
TensorBoard.dev is deprecated, but TensorBoard remains a top choice for experiment tracking. Its deep integration with major ML frameworks, rich visualizations, and flexible local setup contribute to its widespread use by developers and researchers.
Platforms for training100+
of surveyed Python developers work on ML deployment and inference. Interestingly, the most popular tools for this task are in-house solutions.
Platforms for deployment and inference100+
Do you or does your company use tools / platforms for ML workloads in the cloud?
How do computation costs impact your choice of tools or platforms for ML workloads in the cloud?
They are important, but I balance them against performance and features
They are the primary factor; I always seek to minimize costs
They are secondary to other factors like ease of use and integration
Costs are not a major concern
What is your typical monthly budget for cloud-based ML compute resources?
Less than USD 1,000
USD 1,000—5,000
USD 5,000—10,000
USD 10,000—25,000
Over USD 25,000
I'm not sure
of respondents work with big data, with the majority preferring cloud solutions. Among big data tools, PySpark is the most popular, used by 40% of respondents.
Big data tools100+
36%
40%
PySpark
8%
7%
Great Expectations
6%
6%
PyFlink
3%
4%
PyDeequ
5%
4%
Other
50%
49%
None
Solutions used for work with big data100+
Cloud
Self-hosted
Both
None
Development Tools
Operating system100+
Linux
Windows
macOS
BSD
Other
100+
ORMs100+
41%
59%
SQLAlchemy
15%
56%
Django ORM
12%
14%
Raw SQL
10%
14%
SQLModel
All options with less than 2% have been merged into “Other”.
The share of data scientists involved in database development has increased by four percentage points compared to last year.
Could this change be due to the growing use of vector databases in LLM applications?
ORMs100+
34%
39%
SQLAlchemy
25%
26%
Django ORM
13%
12%
Raw SQL
7%
10%
SQLModel
Databases100+
43%
49%
PostgreSQL
34%
37%
SQLite
30%
31%
MySQL
17%
18%
Redis
17%
19%
MongoDB
10%
11%
MariaDB
10%
12%
MS SQL Server
All options with less than 2% have been merged into “Other”.
Continuous integration (CI) systems100+
All options with less than 2% have been merged into “Other”.
Two-thirds of Python developers regularly use continuous integration systems.
GitHub Actions leads the way, followed by GitLab CI/CD and Jenkins/Hudson.
Configuration Management Tools100+
Ansible
A custom solution
Puppet
Chef
Salt
Other
None
Documentation Tools100+
43%
44%
Markdown
25%
29%
Swagger
16%
15%
Sphinx
14%
15%
Postman
13%
11%
Wiki
How do you typically work with a single Python file?100+
I open the entire project that contains the file in an IDE
I use a command-line editor
I open just that one file in an IDE
I use a lightweight text editor
Other
I don't usually need to open or edit individual Python files
Main IDE/Editor
To identify the most popular editors and IDEs, we asked a single-answer question “What is the main editor you use for your current Python development?”.
All options with less than 1% have been merged into “Other”.
Data science vs. Web development
44%
46%
Visual Studio Code
27%
37%
PyCharm
7%
0%
Jupyter Notebook
2%
0%
Spyder
Among VS Code users, the Data Wrangler extension is used by 11%, and 53% take advantage of the IDE’s Jupyter support.
PyCharm
In comparison, Jupyter support is used by 33% of IntelliJ IDEA users and 37% of PyCharm users.
Learn more about Jupyter notebook support and PyCharm’s other features for data professionals.
of surveyed Python developers use additional IDEs or editors alongside their main one, and 42% use three or more simultaneously.
IDEs/Editors used in addition to main IDE/Editor100+
All options with less than 1% have been merged into “Other”.
Number of IDEs/Editors used
1
2
3
>3
Python Packaging
Which of the following tools do you use to isolate Python environments between projects?100+
55%
62%
venv
28%
25%
virtualenv
20%
19%
Conda
18%
18%
Poetry
9%
8%
Pipenv
–
11%
uv
PyCharm
PyCharm provides uv integration, which lets you create new environments from scratch and apply uv to existing environments.
Which tools do you use to manage dependencies?100+
77%
74%
pip
19%
20%
Poetry
19%
18%
Conda
–
12%
uv
9%
8%
Pipenv
9%
9%
pip-tools
What format(s) is your application dependency information stored in?100+
63%
59%
requirements.txt
32%
36%
pyproject.toml
17%
16%
setup.py
11%
12%
I don't store dependency information
Where do you install packages from?100
80%
75%
PyPI
28%
29%
GitHub
16%
16%
Anaconda
14%
14%
A local source
10%
10%
A private Python Package Index
10%
11%
From Linux distribution
10%
11%
An internal mirror of PyPI
Where do you install packages from?100
73%
83%
PyPI
29%
25%
GitHub
27%
6%
Anaconda
15%
10%
A local source
13%
11%
An internal mirror of PyPI
11%
12%
A private Python Package Index
10%
2%
Other Conda channels
of respondents have packaged and published a Python application they developed to a package repository.
Which tools do you use to create packages of your Python libraries?100
Twine
PyPI Publish GitHub Action
Poetry
Hatch
Flit
PDM
Other
How familiar are you with Trusted Publishers?
I've never heard of it
I'm vaguely aware of it
I've tried it, but I don't use it anymore
I currently use it
of surveyed Python developers are working with a monorepo, where multiple packages or services are stored in a single repository, each with its own independently managed dependencies.
Do you use a virtual environment in containers?
Yes
No
Other
I don't use containers for Python development
of respondents build Python binary modules with other languages, primarily C++, C, and Rust. Interestingly, Rust shows an increase of six percentage points compared to last year.
Languages for building binary modules for Python100+
55%
54%
C++
44%
45%
C
27%
33%
Rust
9%
10%
Go
Demographics
Gender
This question was optional.
Age range
What is your country or region?
All options with less than 1% have been merged into “Other”.
Working in a team vs working independently
Working on projects
Employment status
Job roles100+
Company size
Team size
Company industry
All options with less than 1% have been merged into “Other”.
Methodology and Raw Data
Want to dig further into the data? Download the anonymized survey responses and see what you can learn! Share your findings and insights by mentioning @jetbrains and @ThePSF on X with the hashtag #pythondevsurvey.
Before you begin to dissecting this data, please note the following important points:
Once again, on behalf of both the Python Software Foundation and JetBrains, we’d like to thank everyone who took part in this survey. With your help, we’re able to map the landscape of the Python community more accurately!
Contribute to the PSF’s Recurring Giving Campaign. The PSF is a non-profit organization entirely supported by its sponsors, members & the public.
We hope you found our report useful. Share this report with your friends and colleagues!