Here are some of the areas in which Keras compares favorably to existing alternatives.
With around 2.5 million developers as of early 2023, Keras is at the center of a massive community and ecosystem.
You are already constantly interacting with features built with Keras -- it is in use at YouTube, Netflix, Uber, Yelp, Instacart, Zocdoc, Twitter, Square/Block, and many others. Keras is especially popular among startups that place deep learning at the core of their products. Keras is also used in many well-known companies you might not associate with Machine Learning, such as JP Morgan Chase, Orange, and Comcast, and by research units at the likes of NASA, the US DOE, and CERN.
In the 2022 survey "State of Data Science and Machine Learning" by Kaggle, Keras had a 61% adoption rate among Machine Learning developers and Data Scientists [source].
Your Keras models can be easily deployed across a greater range of platforms than any other deep learning API:
Keras is scalable. Using the TensorFlow
DistributionStrategy API, which is supported natively by Keras,
you easily can run your models on large GPU clusters (up to thousands of devices) or an entire TPU pod, representing over one exaFLOPs of computing power.
Keras also has native support for mixed-precision training on the latest NVIDIA GPUs as well as on TPUs, which can offer up to 2x speedup for training and inference.
For more information on data-parallel training, see our guide to multi-GPU & distributed training.
Like you, we know firsthand that building and training a model is only one slice of a machine learning workflow. Keras is built for the real world, and in the real world, a successful model begins with data collection and ends with production deployment.
Keras is at the center of a wide ecosystem of tightly-connected projects that together cover every step of the machine learning workflow, in particular:
Learn more about the Keras ecosystem here.