optimistix
Nonlinear optimisation in JAX and Equinox.
Decision gist · record as of 2026-08-14
Yes, if you work with JAX and need nonlinear solvers. The library is actively maintained, has no known vulnerabilities, installs with low friction, and offers a composable solver interface tailored to JAX's autodiff and GPU/TPU ecosystem. The alpha status means the API may change, but the maintenance signal is strong. Not necessary if you only need first-order gradient optimizers (use Optax instead) or if you are not already in the JAX ecosystem.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.11+; JAX and its dependencies (including jax itself) must be installed and functional.
- Low friction: pure Python wheel with five runtime dependencies (equinox, jax, jaxtyping, lineax, typing-extensions).
- Actively maintained with recent commits and no known vulnerabilities.
License · maintenance · safety
permissive license (permissive) — Apache 2.0 permissive license allows commercial and derivative use with minimal restrictions—include a copy of the license and note any modifications to the source.
last release 2026-02-16 (179 days) · last repo commit 2026-08-11 · 613 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 391,911 downloads/mo, #7,009 on PyPI
Alternatives
Verify before relying
pip install optimistix
import jax.numpy as jnp
import optimistix as optx
y0 = jnp.array(1.)
dt = jnp.array(0.1)
def fn(y, args):
return y0 + jnp.tanh(y) * dt
solver = optx.Newton(rtol=1e-5, atol=1e-5)
sol = optx.fixed_point(fn, solver, y0)
y1 = sol.value- Whether solver convergence guarantees or failure modes are documented for each algorithm.
- Performance comparison with other JAX optimization libraries or non-JAX solvers.
- Supported problem sizes and scalability limits on GPU/TPU.
What it is and what it does
Optimistix is a JAX library for solving nonlinear problems: finding roots, minimizing functions, computing fixed points, and solving least-squares problems. It wraps these problem types into a unified interface so you can, for example, convert a root-finding problem to a least-squares problem and solve it with a minimization algorithm. The library is built around modular solvers—you can compose descent paths (like dogleg), update rules (like trust region), and quadratic models (like BFGS) to construct custom optimizers.
Because it's built on JAX, Optimistix inherits autodifferentiation, automatic parallelization, and GPU/TPU support. It works with PyTree-based state, integrates with the Equinox neural-network library, and can interoperate with Optax optimizers. The library is in alpha (Development Status 3) and actively maintained, with documentation available online.
Use it for
- Solve implicit differential equations (e.g., implicit Euler) by finding fixed points of the update map.
- Minimize loss functions in neural networks or scientific models using custom composed solvers.
- Find roots of nonlinear systems for physics simulations or engineering problems.
- Solve least-squares problems for parameter fitting or inverse problems.
- Combine multiple solver types in a single JAX-compiled pipeline with autodiff.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you work with JAX and need nonlinear solvers.
The library is actively maintained, has no known vulnerabilities, installs with low friction, and offers a composable solver interface tailored to JAX's autodiff and GPU/TPU ecosystem. The alpha status means the API may change, but the maintenance signal is strong. Not necessary if you only need first-order gradient optimizers (use Optax instead) or if you are not already in the JAX ecosystem.
Install
optimistix on PyPI
Before you install
Low friction: pure Python wheel with five runtime dependencies (equinox, jax, jaxtyping, lineax, typing-extensions). Actively maintained with recent commits and no known vulnerabilities. Requires Python 3.11+.
Requires Python 3.11+; JAX and its dependencies (including jax itself) must be installed and functional.
License in practice
Apache 2.0 permissive license allows commercial and derivative use with minimal restrictions—include a copy of the license and note any modifications to the source.
Quickstart
pip install optimistix
import jax.numpy as jnp
import optimistix as optx
y0 = jnp.array(1.)
dt = jnp.array(0.1)
def fn(y, args):
return y0 + jnp.tanh(y) * dt
solver = optx.Newton(rtol=1e-5, atol=1e-5)
sol = optx.fixed_point(fn, solver, y0)
y1 = sol.value
Verify before relying
- Whether solver convergence guarantees or failure modes are documented for each algorithm.
- Performance comparison with other JAX optimization libraries or non-JAX solvers.
- Supported problem sizes and scalability limits on GPU/TPU.
Package facts
| License | permissive license permissive |
| Python support | Supports the current Python release ~=3.11 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 5 packagesequinoxjaxjaxtypinglineaxtyping-extensions |
| Maintenance | Actively maintained 179 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 391,911 / month, #7,009 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 3 - AlphaIntended Audience :: DevelopersIntended Audience :: Financial and Insurance IndustryIntended Audience :: Information TechnologyIntended Audience :: Science/ResearchLicense :: OSI Approved :: Apache Software LicenseNatural Language :: EnglishProgramming Language :: Python :: 3Topic :: Scientific/Engineering :: Artificial IntelligenceTopic :: Scientific/Engineering :: Information AnalysisTopic :: Scientific/Engineering :: Mathematics |
Evidence: optimistix-0.1.0-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “nonlinear solvers jax”
- optimistixOptimistix provides nonlinear solvers for root finding, minimization,…
- klujaxSolves sparse linear systems in JAX using the KLU algorithm from…
- diffraxDiffrax provides numerical solvers for ordinary, stochastic, and…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also diffrax · lineax · optax · directsearch · torchdiffeq · ott-jax · xpress · google-tunix · nvidia-cusolver