Gaussian elimination step by step
About this app
Perform row operations with exact fractions, reach echelon form and back-substitute. Explore unique, absent or infinitely many solutions, matrix inverses and intersecting planes.
With explain mode: an AI tutor walks you through the app, points out what matters in the scene and answers your questions.
The method
A system of linear equations is written as an augmented matrix (A | b): coefficients on the left, right-hand side on the right. Three row operations leave the solution set unchanged: swapping two rows, multiplying a row by k ≠ 0, and adding a multiple of one row to another. The goal is row echelon form, with zeros below every pivot. Back substitution then gives the solution. A zero row 0 = 0 means a free parameter; a row 0 = c with c ≠ 0 is a contradiction, so there is no solution.
What you see
The augmented matrix is computed with exact fractions, so there are no rounding errors. Every operation is logged, and a status line names the current pivot and the next step. The Solution panel shows back substitution and the solution case: exactly one, none or infinitely many solutions with the number of parameters. With three unknowns each equation is a plane in space; at every step one plane tilts while the common intersection point stays put. Applied to (A | I), the same procedure produces the inverse A⁻¹ on the right.
Things to try
Start with an example such as “Machines 3×3”, “Raw materials 4×4”, “No solution” or “Infinitely many” and step forward with “Next step” or let the app run automatically. You can also choose operations yourself: target row, source row and factor (fractions like −3/2 work), then apply. Under “Custom system” you type in your own system, one line per equation. In inverse mode you see when Gauss-Jordan reaches the identity on the left, and when a zero row shows that A is not invertible.
Frequently asked questions
- How does Gaussian elimination work?
- You bring the augmented matrix (A | b) to row echelon form using row operations: in each column you pick a nonzero pivot and create zeros below it by subtracting suitable multiples of the pivot row. Then you solve from the last equation upwards (back substitution). The row operations do not change the solution set.
- How can you tell if a linear system has no solution or infinitely many?
- Look at the echelon form. A row of the form 0 = c with c ≠ 0 is a contradiction, so there is no solution. If there is no contradiction but fewer pivots than unknowns, free parameters remain and there are infinitely many solutions. Otherwise the solution is unique.
- How do you find an inverse matrix with Gauss-Jordan elimination?
- Write A next to the identity matrix as (A | I) and apply row operations until the left block is I. The right block is then A⁻¹. If a zero row appears on the left, A does not have full rank and is not invertible.
- What can I do in the Gaussian elimination app?
- You solve example systems or your own step by step or automatically, with exact fractions and a log of every operation. With three unknowns you see the equations as planes in space, and inverse mode computes A⁻¹ via (A | I).
Subject: Linear algebra
More from Linear algebra
- Eigenvectors
- Equations as scales
- Linear combination & linear independence (3D)
- Matrix × vector
- Matrizen rechnen: Produkt, Inverse, Transponierte
- Matrices like numbers: identity, inverse, A·x = b
- Matrix multiplication
- Horse and rail: the dot product as work
A VisuApp by heyprof: interactive, free in your browser, no sign-up. What is a VisuApp? · All apps