> In 2019, the war for ML frameworks has two remaining main contenders: PyTorch and TensorFlow. My analysis suggests that researchers are abandoning TensorFlow and flocking to PyTorch in droves.

Seems they were pretty spot on! https://trends.google.com/trends/explore?date=all&q=pytorch,...

But to be fair, it was kind of obvious around ~2023 without having to look at metrics/data, you just had to look at what the researchers publishing novel research used.

Any similar articles that are a bit more up to date, maybe even for 2025?