# SciML/NeuralPDE.jl

Physics-Informed Neural Networks (PINN) Solvers of (Partial) Differential Equations for Scientific Machine Learning (SciML) accelerated simulation

Repository: https://github.com/SciML/NeuralPDE.jl
Canonical: https://ross.abutalabs.com/products/neuralpdejl
Homepage: https://docs.sciml.ai/NeuralPDE/stable/
Language: Julia
License: NOASSERTION
License Family: other
Topics: differential-equations, differentialequations, neural-differential-equations, scientific-ml, scientific-ai, partial-differential-equations, ordinary-differential-equations, neural-network, neural-networks, ode, pde, sciml, scientific-machine-learning, pinn, machine-learning
Last push: 2026-08-26T17:29:38+00:00

## Health v2 (maintenance only)
Score: 99/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 98, longevity 100
- inputs: {"age_days": 3459, "days_push": 7, "days_rel": 15, "gap_med": 20, "n_releases_24m": 16}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1220, forks 249 (observed 2026-08-28T04:04:01.859005+00:00)

## What it is
NeuralPDE.jl is a Julia library of physics-informed neural network (PINN) solvers for ordinary, stochastic, and partial differential equations. It automatically constructs physics-informed loss functions from a symbolic interface and integrates with Flux.jl, Lux.jl, and NeuralOperators.jl for GPU-accelerated scientific machine learning.

## Use cases
- solve partial differential equations with physics-informed neural networks
- fit a neural network to both physical laws and experimental data
- solve ODEs and SDEs using neural network solvers
- train PINNs with automatic loss construction from symbolic equations
- combine DeepONet or Fourier neural operators with physics-informed losses
- accelerate PDE simulation with GPU-powered training

## When to choose
- you work in Julia and need PINN-based differential equation solvers
- you want to mix equation solving with data fitting in a scientific machine learning workflow
- you need automated physics-informed loss functions from a high-level symbolic interface
- you want neural operator methods like DeepONets combined with physics constraints

## When to avoid
- you need classical finite-element or finite-difference PDE solvers rather than neural approaches
- you work primarily in Python rather than Julia
- you need a lightweight tool without a machine learning stack
- your equations are simple enough that classical numerical solvers are faster and more accurate

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, simulation, deep-learning
- domain: machine-learning, simulation
- platform: cross-platform
- tags: pinn, physics-informed-neural-networks, pde-solver, ode-solver, sciml, neural-operators, flux, lux, solver, scientific-machine-learning, differential-equations, julia, gpu

## Member repositories
- SciML/NeuralPDE.jl (main) score 99

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:01.859005+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 2026-08-30T06:15:56.173795+00:00, confidence not recorded.
  - readme: https://github.com/SciML/NeuralPDE.jl (fetched 2026-08-28T04:04:01.859005+00:00, sha 606651d24ea2)
  - homepage: https://docs.sciml.ai/NeuralPDE/stable/ (fetched 2026-08-29T12:24:13.278106+00:00, sha 22da17c23c08)
  - site_page: https://docs.sciml.ai/DiffEqDocs (fetched 2026-08-29T12:24:13.288124+00:00, sha 82260832e07c)
  - site_page: https://docs.sciml.ai/DiffEqDevDocs (fetched 2026-08-29T12:24:13.289960+00:00, sha db0de492634b)
- Data as of 2026-08-30T08:39:29.467469+00:00.