JAX Comparisons

Comments primarily compare the discussed ML library or tool to Google's JAX, focusing on differences in performance, JIT compilation, functional style, and features relative to PyTorch, TensorFlow, and Numba.

📉 Falling 0.4x AI & Machine Learning
1,731
Comments
17
Years Active
5
Top Authors
#3587
Topic ID

Activity Over Time

2010
1
2011
2
2012
3
2013
3
2014
8
2015
11
2016
18
2017
88
2018
109
2019
121
2020
200
2021
199
2022
212
2023
312
2024
243
2025
189
2026
12

Keywords

CPU JIT S3 TensorFlow PyTorch AWS CSV TPU youtube.com NumPy jax pytorch numpy jit tensorflow gpu transformations cuda backends neck

Sample Comments

tanilama Apr 16, 2019 View on HN

How does this compare against Google's JAX?

kavalg Mar 12, 2023 View on HN

How different is this from JAX in practice?

chazeon Dec 2, 2022 View on HN

How is PyTorch compares to JAX and its stack?

verdverm Oct 27, 2021 View on HN

This seems to have similarities to JAX, ie defining as functions and getting some DAGs and auto vectorization

KKKKkkkk1 Jan 10, 2025 View on HN

Which structural limits of TF2 and PyTorch were fixed via the Jax ecosystem?

nerpderp82 Oct 26, 2022 View on HN

You should give JAX a go.https://github.com/google/jax

galaxyquanta Apr 20, 2023 View on HN

What does this mean for JAX (light-weight ML library from Google Brain) vs Tensorflow (from Deepmind)?

fho Feb 28, 2024 View on HN

Funfact: you can probably JIT compile that using JAX for an easy performance gain.

6gvONxR4sf7o Jun 14, 2024 View on HN

Don’t forget JAX! It’s my preferred library for “i want to write numpy but want it to run on gpu/tpu with auto diff etc”

zitterbewegung Feb 12, 2022 View on HN

Is this in direct competition with JAX from google ?