Matrix Rank Calculator — Row-Echelon Form

📋

Walk-through

How to Use This Calculator

4 steps

1

Choose the matrix dimensions

Pick the number of rows and columns (up to 5×5 each, independently) from the two selectors. The entry grid resizes to match, and a worked full-rank example loads automatically so you can see the method in action right away.

2

Enter your matrix values

Click any cell and type a number — positive, negative, or decimal. The rank, row-echelon form, and full elimination steps all update instantly as you edit; there's no Calculate button to press.

3

Read the rank

The result card shows the rank as soon as the matrix is entered, along with the maximum rank possible for that shape (the smaller of rows and columns). The Rank tab breaks this down into dimensions, rank, max possible rank, and a Full Rank / Rank-Deficient status.

4

Check the row-echelon form and steps

Switch to the Row-Echelon Form tab to see the matrix after Gaussian elimination, or the Steps tab to watch every row swap and elimination substep that produced it, with the row being changed highlighted at each step.

⚡

Reference

Formula & Methodology

2 formulas

Rank via row-echelon reduction

rank(A) = number of non-zero rows in REF(A)

Gaussian elimination with partial pivoting reduces a matrix to row-echelon form (REF): at each step, the row with the largest absolute value in the current pivot column is swapped to the top, then every row below it has a multiple of the pivot row subtracted to zero out that column. Repeating this for each column leaves a staircase pattern of pivots. The rank is simply the count of rows that are not entirely zero once this process finishes.

Bounds on the rank

0 ≤ rank(A) ≤ min(rows, columns)

A matrix's rank can never exceed the smaller of its row count and column count — this is the "max possible rank" shown alongside the result. When rank(A) equals that maximum, the matrix is full rank. When it's smaller, at least one row is a linear combination of the others (linearly dependent), and row reduction collapses it to an all-zero row.

📖

Glossary

Key Terms Explained

7 terms

Rank The number of linearly independent rows in a matrix, equal to the number of non-zero rows after reducing it to row-echelon form. It measures how much independent information the matrix actually contains.

Row-echelon form (REF) A staircase-shaped version of a matrix produced by Gaussian elimination, where each row's leading (leftmost non-zero) entry sits strictly to the right of the row above it, and every entry below a leading entry is zero.

Gaussian elimination A systematic method for reducing a matrix to row-echelon form using three row operations: swapping two rows, scaling a row, and adding a multiple of one row to another. This calculator uses partial pivoting (selecting the largest available entry in each column) to keep the elimination numerically stable.

Linearly independent A set of rows is linearly independent when none of them can be written as a combination (sum of scaled copies) of the others. A matrix with all rows linearly independent has full rank.

Pivot The first non-zero entry used to eliminate the entries below it in the same column during Gaussian elimination. The number of pivots found equals the rank.

Non-zero row A row of the row-echelon form that contains at least one non-zero entry. Counting these rows (after eliminating floating-point noise near zero) gives the rank.

Full rank A matrix is full rank when its rank equals min(rows, columns) — the maximum possible for its shape. A full-rank square matrix is always invertible; a rank-deficient one is not.

👥

Scenarios

Real-World Examples

3 worked examples

🔢

Full-rank matrix

A = [[2, 1, 1], [1, 3, 2], [1, 0, 0]] (3×3)

Matrix A [[2, 1, 1], [1, 3, 2], [1, 0, 0]]

Gaussian elimination reduces this to a row-echelon form with three non-zero rows, so rank(A) = 3 — the maximum possible for a 3×3 matrix. All three rows are linearly independent, and the matrix is full rank (and therefore invertible).

⚠️

Rank-deficient matrix

A = [[1, 2, 3], [2, 4, 6], [1, 1, 1]] (3×3)

Matrix A [[1, 2, 3], [2, 4, 6], [1, 1, 1]]

Row 2 is exactly twice Row 1, so eliminating it produces an all-zero row. Only two non-zero rows remain in the row-echelon form, giving rank(A) = 2 — one short of the maximum of 3. The calculator flags this as rank-deficient and names the dependent row.

🔢

Non-square matrix

A = [[1, 2, 3], [4, 5, 6]] (2×3)

Matrix A [[1, 2, 3], [4, 5, 6]]

For a 2×3 matrix the maximum possible rank is min(2, 3) = 2. Row reduction leaves two non-zero rows, so rank(A) = 2 — full rank for this shape, even though the matrix isn't square.

📄

Deep Dive

Understanding the Matrix Rank Calculator

3 chapters · 4 min read

The rank of a matrix tells you how much independent information it actually holds — a concept that underlies whether a system of linear equations has a unique solution, infinitely many, or none at all. This calculator finds the rank of any matrix up to 5×5 (square or rectangular) by reducing it to row-echelon form via Gaussian elimination, the same method taught in an introductory linear algebra course.

How row-echelon reduction reveals the rank

Gaussian elimination works column by column. For each column, the calculator picks the row with the largest absolute value in that column (partial pivoting, which keeps the arithmetic numerically stable) and swaps it to the top of the remaining rows if needed. It then subtracts a multiple of that pivot row from every row below it, zeroing out the entries beneath the pivot. Repeating this for each column produces a staircase-shaped row-echelon form.

Once the matrix is in row-echelon form, the rank is just the number of rows that aren't entirely zero. Any row that became all zeros during elimination represents a row of the original matrix that was a linear combination of the others — it carried no new information.

Inputs and what they mean

Choose the number of rows and columns independently (1 to 5 each), then fill in the grid with any real numbers. The matrix does not need to be square — rectangular matrices are fully supported, and the calculator automatically caps the maximum possible rank at min(rows, columns). Use the preset buttons to load a full-rank example, a rank-deficient example with dependent rows, a non-square example, or the identity matrix.

Limits and edge cases

Rank is always a whole number between 0 (only for the all-zero matrix) and min(rows, columns). Floating-point arithmetic can leave tiny non-zero residues (like 1e-15) where a value should be exactly zero — the calculator treats anything smaller than a small tolerance as zero so these numerical artifacts don't inflate the rank. This calculator supports matrices up to 5×5; for full matrix inversion use the dedicated Matrix Inverse Calculator, and for solving a system of linear equations directly use the System of Equations Calculator.

❓

Questions

Frequently Asked Questions

6 questions

What's the formula for matrix rank?+

Reduce the matrix to row-echelon form using Gaussian elimination, then count the non-zero rows — that count is the rank. There's no closed-form algebraic formula; rank is defined by this reduction process (or equivalently, by the largest number of linearly independent rows or columns).

What does it mean for a matrix to have full rank?+

A matrix has full rank when rank(A) equals min(rows, columns) — the maximum possible for its shape. A full-rank square matrix is always invertible. A matrix with fewer rows than columns has full row rank when its rank equals the number of rows.

What does it mean if rows are linearly dependent?+

It means the matrix's rank is less than min(rows, columns) — at least one row can be written as a combination of the others (for example, one row being an exact multiple of another). Gaussian elimination reveals this by collapsing the dependent row to all zeros.

What is matrix rank actually used for?+

Rank determines whether a system of linear equations Ax = b has a unique solution, infinitely many solutions, or none — it's central to solvability analysis. It's also used to check whether a matrix is invertible, to measure the dimension of a vector space spanned by a set of vectors, and in data science to detect redundant features.

What is Gaussian elimination?+

Gaussian elimination is a step-by-step method that uses three row operations — swapping two rows, scaling a row, and subtracting a multiple of one row from another — to reduce a matrix to row-echelon form. This calculator uses partial pivoting, always eliminating using the largest available entry in each column, for better numerical stability.

Can this calculator find the rank of a non-square matrix?+

Yes. Rows and columns are set independently (each from 1 to 5), so rectangular matrices are fully supported. For a non-square matrix the maximum possible rank is min(rows, columns), which the calculator displays alongside the computed rank.

Download a one-page PDF of your numbers instantly. Add your email to also get our occasional calculator tips — no spam, unsubscribe anytime. Privacy.

Rank