How to solve a system of equations with RREF

By Remetric Labs · Last reviewed October 26, 2026

What RREF tells you

Reduced row echelon form exposes pivots: a leading 1 in every non-zero row, zeros above and below each pivot. Read the augmented matrix left to right - each pivot column is a basic variable, each non-pivot column is free. Try each system below in the RREF calculator, which shows every elimination step.

Case 1: one solution

System: x + y = 3 and 2x − y = 0. Augmented matrix [[1,1|3],[2,−1|0]]. Subtract 2×R1 from R2 → [[1,1|3],[0,−3|−6]]; scale R2 by −1⁄3 → [[1,1|3],[0,1|2]]; subtract R2 from R1 → [[1,0|1],[0,1|2]]. Rank 2 with two unknowns: x = 1, y = 2. Check: 1 + 2 = 3 and 2 − 2 = 0.

Case 2: infinitely many solutions

System: x + 2y = 4 and 2x + 4y = 8 (the second equation is twice the first). RREF gives [[1,2|4],[0,0|0]] - one pivot, one free variable. Let y = t; then x = 4 − 2t. Every point on that line satisfies both equations, so report the parametric family, not a single pair.

Case 3: no solution

System: x + y = 1 and x + y = 3. RREF gives [[1,1|1],[0,0|1]] - the bottom row reads 0 = 1, a contradiction. The lines are parallel and never meet. A zero row with a non-zero right-hand side always means inconsistency; a zero row of all zeros means dependence (case 2).

Limitations

RREF with floating-point arithmetic can show 0.9999999 instead of 1 on ill-conditioned matrices - round display values before interpreting. Over-determined systems (more equations than unknowns) usually have no exact solution; consider least squares instead. Related: Matrix Multiplication, Determinant, Inverse Matrix.