AI-Powered Materials Discovery

Discover Materials
Accelerated

MatForge combines surrogate models, active learning, and Pareto optimization to find optimal materials 100x faster than brute-force search. No GPU required.

Free to explore · No credit card required · 11 material domains

3D Structure
Interactive
100x
Faster Discovery
16
Material Domains
<1s
Per Evaluation
0
GPU Required
Capabilities

Everything You Need for Materials Discovery

A complete platform combining physics-based evaluation, machine learning surrogates, and intelligent optimization.

Surrogate Models

NumPy-only MLP neural network with MC Dropout uncertainty. No GPU needed - runs anywhere.

Active Learning

Smart sampling with MaxUncertainty, Expected Improvement, and UCB acquisition functions.

Pareto Optimization

NSGA-II multi-objective optimization with CMA-ES on the surrogate surface.

11 Domain Plugins

Water, battery, solar, CO2, catalyst, hydrogen, construction, bio, agri, electronics, textile.

100x Faster

Replace expensive simulations with surrogate predictions. Evaluate thousands in seconds.

Real-time Dashboard

Live campaign progress with 3D visualizations, Pareto plots, and convergence tracking.

Workflow

How It Works

01

Define

Write a YAML material definition: parameters, objectives, constraints, and physics equations.

02

Launch

Start a campaign. The engine samples initial materials and trains a surrogate model.

03

Optimize

CMA-ES finds optimal candidates on the surrogate. Active learning picks the most informative.

04

Discover

Pareto-optimal materials emerge. Export recipes, visualize trade-offs, iterate.

Material Domains

16 Domains, Infinite Possibilities

Each domain includes physics equations, YAML templates, and pre-configured optimization parameters.

Water Filtration

PFOS rejection membranes

Battery Materials

Next-gen energy storage

Solar Cells

Perovskite photovoltaics

CO2 Capture

Carbon capture sorbents

Catalysis

Reaction optimization

Hydrogen Storage

H2 storage materials

Construction

Low-carbon concrete

Biomaterials

Biocompatible scaffolds

Agriculture

Controlled-release fertilizers

Electronics

Semiconductor materials

Smart Textiles

Responsive fabric composites

Thermoelectrics

Heat-to-electricity conversion

Superconductors

High-Tc superconducting materials

Polymers

Sustainable polymer design

Coatings

Protective thin film coatings

Ceramics

High-performance ceramics

Why MatForge

Built for Scientists

Hours, Not Months

Evaluate 10,000 candidates where you could only test 50.

Physics-Grounded

Equations from domain experts. Not just statistics.

Zero Setup

No GPU, no cloud credentials, no Docker. pip install and go.

Open & Extensible

Write your own plugins. Bring your own evaluator.

name: PFOS Rejection Membrane
domain: water

parameters:
  - name: pore_diameter
    range: [0.5, 5.0]
    unit: nm
  - name: active_layer_thickness
    range: [50.0, 500.0]
    unit: nm

objectives:
  - name: pfos_rejection
    direction: maximize
    equation: "water:pfos_rejection"
  - name: permeability
    direction: maximize
    equation: "water:permeability"

active_learning:
  initial_samples: 20
  samples_per_round: 10
  acquisition: max_uncertainty
material.yaml

Ready to Forge Your Next Material?

Join researchers using MatForge to discover novel materials for water, energy, construction, and beyond.