jax-ml/jax
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more observed · 2026-10-18
Health v2 · maintenance only
98/100
- Activity 100
- Release rhythm 95
- Longevity 100
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-10-21. Adoption (stars, forks) is never an input.
- gap_med: 28
- age_days: 2917
- days_rel: 33
- days_push: 3
- n_releases_24m: 28
Adoption not part of the score
36460 stars · 3862 forks observed · 2026-10-18
What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-28, confidence 0.97
JAX is a Python library that provides composable transformations such as automatic differentiation, JIT compilation, and vectorization for NumPy code. It leverages XLA to run efficiently on CPUs, GPUs, and TPUs.
Use cases
- compute gradients for neural network training
- accelerate scientific NumPy code on GPUs or TPUs
- vectorize batch operations with vmap for fast per‑example processing
- compile performance‑critical functions with jit for low‑latency inference
- experiment with custom kernels using the Pallas extension
When to choose
- You need high‑performance numerical computing with automatic differentiation.
- Your workload must run on multiple accelerator backends (CPU, GPU, TPU).
- You want a NumPy‑compatible API with advanced transformation capabilities.
When to avoid
- You require a fully‑featured deep‑learning framework with built‑in layers and optimizers.
- Your project depends on static‑graph frameworks that do not support JAX's functional style.
- You need extensive ecosystem tools (e.g., model serving, visualization) that are not provided by JAX.
Facets
library · maturity active
machine-learning deep-learning compiler artificial-intelligence machine-learning deep-learning data-science windows python jit grad vmap xla accelerator numpy-compatible high-performance vectorization automatic-differentiation numerical-computing linux macos cpu gpu tpu
10 sources
- readme: https://github.com/jax-ml/jax · fetched 2026-10-18 · 7c101dab4b0f
- homepage: https://docs.jax.dev · fetched 2026-09-14 · 7a112cd10fa7
- site_page: https://docs.jax.dev/en/latest/pallas/quickstart.html · fetched 2026-09-14 · 392db64cd25d
- site_page: https://docs.jax.dev/en/latest/pallas/CHANGELOG.html · fetched 2026-09-14 · d1860f4d10b4
- site_page: https://docs.jax.dev/en/latest/about.html · fetched 2026-09-14 · dce08330eb76
- site_page: https://docs.jax.dev/en/latest/faq.html · fetched 2026-09-14 · c4cb7ecaba18
- site_page: https://docs.jax.dev/en/latest/installation.html · fetched 2026-09-14 · 271ba08ebebd
- site_page: https://docs.jax.dev/en/latest/pallas/tpu/quickstart.html · fetched 2026-09-14 · 76c783c3a037
- site_page: https://docs.jax.dev/en/latest/pallas/gpu/quickstart.html · fetched 2026-09-14 · bdc9e6b9b200
- site_page: https://docs.jax.dev/en/latest/changelog.html · fetched 2026-09-14 · 955198623ea0
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| jax-ml/jax | main | 98 |
For agents
Data as of 2026-10-19T08:05:51.212177+00:00 · Report a problem