# google/jaxopt

Hardware accelerated, batchable and differentiable optimizers in JAX.

Repository: https://github.com/google/jaxopt
Canonical: https://ross.abutalabs.com/products/jaxopt
Homepage: https://jaxopt.github.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: jax, optimization, deep-learning, differentiable-programming, bi-level
Last push: 2026-06-04T21:38:27+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 85, release rhythm 40, longevity 100
- inputs: {"age_days": 1878, "days_push": 90, "days_rel": 506, "gap_med": 4, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1059, forks 76 (observed 2026-08-28T04:03:25.565928+00:00)

## What it is
JAXopt is a Python library providing hardware-accelerated, batchable, and differentiable optimizers built on JAX, including implicit differentiation of optimization problem solutions. The project is no longer maintained, with many features ported into Optax.

## Use cases
- differentiate through optimization problems in JAX
- run optimizers on GPU and TPU
- vectorize many instances of an optimization problem with vmap
- implicit differentiation of fixed points and constrained problems
- bilevel optimization and hyperparameter tuning
- root finding and solvers in JAX pipelines

## When to choose
- you need differentiable or bilevel optimization in JAX and are fine using an unmaintained library
- you want implicit differentiation of optimization solutions
- you need batched solvers accelerated on GPU/TPU

## When to avoid
- starting a new project - use Optax or other JAX alternatives instead
- you need maintained software with ongoing support
- you need optimizers outside the JAX ecosystem

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, math
- domain: machine-learning, deep-learning
- platform: python
- tags: jax, optimization, implicit-differentiation, bilevel-optimization, root-finding, differentiable-programming, algorithms, gpu

## Member repositories
- google/jaxopt (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:25.565928+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:57:14.316062+00:00, confidence not recorded.
  - readme: https://github.com/google/jaxopt (fetched 2026-08-28T04:03:25.565928+00:00, sha 2b9bdd3865d7)
  - homepage: https://jaxopt.github.io (fetched 2026-08-29T12:59:18.741282+00:00, sha 9b0494b66f1b)
  - registry_pypi: https://pypi.org/pypi/jaxopt/json (fetched 2026-08-29T12:59:18.750346+00:00, sha 54203dfc1dba)
- Data as of 2026-08-30T08:39:29.467469+00:00.