Keras

When to Use

• Portability: Write once, run on any framework.
• Simplicity:

  model.fit()

  is still the cleanest API in the industry.
• XLA: Keras 3 enables XLA compilation on all backends by default.

Core Concepts

Backend Agnostic

os.environ["KERAS_BACKEND"] = "jax"

Functional API

x = Dense()(inputs)

Keras Core (

keras.ops

)

Best Practices (2025)

• Use Keras 3: Migrate from

  tf.keras

  .
• Use JAX backend: For fastest training on TPUs/GPUs.
• Use PyTorch backend: If you need to integrate into a larger PyTorch codebase.

• Don't mix

  tf.*

  ops: Use

  keras.ops.*

  to remain framework-agnostic.

References

• Keras Documentation