RASON
RASON (RESTful Analytic Solver Object Notation) is a modeling language and analytics platform embedded in JSON and delivered via a REST API that makes it simple to create, test, solve, and deploy decision services powered by advanced analytic models directly into applications. It lets users define optimization, simulation, forecasting, machine learning, and business rules/decision tables using a high-level language that integrates naturally with JavaScript and RESTful workflows, making analytic models easy to embed into web or mobile apps and scale in the cloud. RASON supports a wide range of analytic capabilities, including linear and mixed-integer optimization, convex and nonlinear programming, Monte Carlo simulation with multiple distributions and stochastic programming methods, and predictive models such as regression, clustering, neural networks, and ensembles, plus DMN-compliant decision tables for business logic.
Learn more
Tidy3D
Tidy3D is Flexcompute's ultrafast electromagnetic (EM) solver. The solver is based on the finite-difference time-domain (FDTD) method. Thanks to the highly optimized co-design of software and hardware, Tidy3D runs simulations orders of magnitude faster than other EM solvers on the market. With the lightning fast speed, you can also solve problems hundreds of wavelength in size, which is often not feasible with conventional approaches.
Learn more
Solver SDK
Use optimization and simulation models in your desktop, Web or mobile application. Use the same high-level objects (like Problem, Solver, Variable and Function), collections, properties and methods across different programming languages. The same object-oriented API is exposed "over the wire" through Web Services WS-* standards to remote clients in PHP, JavaScript, C# and other languages. Procedural languages can use conventional calls that correspond naturally to the properties and methods of the Object-Oriented API. Linear and quadratic programming, mixed-integer programming, smooth nonlinear optimization, global optimization, and non-smooth evolutionary and tabu search are all included. The world's best optimizers, from Gurobi™, XPRESS™ and MOSEK™ for linear, quadratic and conic models to KNITRO™, SQP and GRG methods for nonlinear models "plug into" Solver SDK. Easily create a sparse DoubleMatrix object with 1 million rows and columns.
Learn more
SSPLAX
SSPLAX is a no-code visual platform for modeling and solving constrained operational decisions. Users build models as networks, with stages, resources and outcomes as nodes and routes, processes and decisions as edges. They then define an objective and constraints such as capacity, budget, labor, materials and policy. SSPLAX compiles the model into a linear or mixed-integer optimization problem, solves it, and explains the result: which constraints limit the outcome, why a target is infeasible, which small change would improve it most, and whether added budget or capacity creates value or merely shifts the bottleneck. It also supports assignment, blending and formulation, capacity allocation, and project-portfolio selection. Users can compare scenarios and test how recommendations change under different assumptions. Worked examples cover manufacturing, life sciences, AI infrastructure and decarbonization. SSPLAX is domain-agnostic, browser-based and requires no setup.
Learn more