If you're looking to leverage the insane power of modern GPUs for data science and ML, you might think you'll need to use some low-level programming language such as C++. But the folks over at NVIDIA have been hard at work building Python SDKs which provide nearly native level of performance when doing Pythonic GPU programming. Bryce Adelstein Lelbach is here to tell us about programming your GPU in pure Python.

Episode sponsors

Posit
Agntcy
Talk Python Courses

Links from the show

Bryce Adelstein Lelbach on Twitter: @blelbach

Episode Deep Dive write up: talkpython.fm/blog

NVIDIA CUDA Python API: github.com
Numba (JIT Compiler for Python): numba.pydata.org
Applied Data Science Podcast: adspthepodcast.com
NVIDIA Accelerated Computing Hub: github.com
NVIDIA CUDA Python Math API Documentation: docs.nvidia.com
CUDA Cooperative Groups (CCCL): nvidia.github.io
Numba CUDA User Guide: nvidia.github.io
CUDA Python Core API: nvidia.github.io
Numba (JIT Compiler for Python): numba.pydata.org
NVIDIA’s First Desktop AI PC ($3,000): arstechnica.com
Google Colab: colab.research.google.com
Compiler Explorer (“Godbolt”): godbolt.org
CuPy: github.com
RAPIDS User Guide: docs.rapids.ai
Watch this episode on YouTube: youtube.com
Episode #509 deep-dive: talkpython.fm/509
Episode transcripts: talkpython.fm

--- Stay in touch with us ---
Subscribe to Talk Python on YouTube: youtube.com
Talk Python on Bluesky: @talkpython.fm at bsky.app
Talk Python on Mastodon: talkpython
Michael on Bluesky: @mkennedy.codes at bsky.app
Michael on Mastodon: mkennedy

#509: GPU Programming in Pure Python

Talk Python To Me