Last verified · v1.0

Calculator · math

Matrix Rank Calculator (3×3)

Compute the rank of a 3×3 matrix via Gaussian elimination. Enter all 9 entries and get the rank — 0, 1, 2, or 3 — instantly.

FreeInstantNo signupOpen source

Inputs

Row 1, Col 1 (a₁₁)

Row 1, Col 2 (a₁₂)

Row 1, Col 3 (a₁₃)

Row 2, Col 1 (a₂₁)

Row 2, Col 2 (a₂₂)

Row 2, Col 3 (a₂₃)

Row 3, Col 1 (a₃₁)

Row 3, Col 2 (a₃₂)

Row 3, Col 3 (a₃₃)

Matrix Rank

—

Get a plain-English breakdown of your result with practical next steps.

Matrix Rank—

The formula

How the
result is
computed.

What Is Matrix Rank?

The rank of a matrix A, written rank(A), measures the maximum number of linearly independent rows — equivalently, the maximum number of linearly independent columns — contained in the matrix. A foundational result in linear algebra states that row rank and column rank are always equal. Formally, rank(A) equals the dimension of the column space of A:

rank(A) = dim(col(A)) = number of non-zero rows in RREF(A)

For a 3×3 matrix, rank is an integer between 0 and 3. A rank of 3 means the matrix is full rank: all three rows are linearly independent, the determinant is non-zero, and the matrix is invertible. A rank of 2 or lower signals linear dependence among the rows and a singular (non-invertible) matrix whose determinant equals zero.

The 3×3 Matrix Variables

A 3×3 matrix A contains nine scalar entries, each identified by its row index and column index. Enter these nine values into the calculator fields:

Row 1: a₁₁ (row 1, column 1), a₁₂ (row 1, column 2), a₁₃ (row 1, column 3)
Row 2: a₂₁ (row 2, column 1), a₂₂ (row 2, column 2), a₂₃ (row 2, column 3)
Row 3: a₃₁ (row 3, column 1), a₃₂ (row 3, column 2), a₃₃ (row 3, column 3)

These nine values completely define the matrix and its rank. Even changing a single entry from zero to non-zero can increase the rank, so accurate input of all entries is essential for a correct result.

Computing Rank via Gaussian Elimination

The standard algorithm for matrix rank uses Gaussian elimination to transform A into Reduced Row Echelon Form (RREF). Three elementary row operations — all of which preserve rank — are applied repeatedly:

Swap any two rows
Multiply any row by a non-zero scalar
Add any scalar multiple of one row to another row

After reduction, the RREF of A contains rows with leading 1s (pivots) and rows consisting entirely of zeros. The rank equals the number of pivot rows. This result is formally established by the Rank Theorem (Georgia Tech Interactive Linear Algebra). For a 3×3 matrix, at most three elimination passes are needed to reach RREF.

Worked Example: Rank-Deficient Matrix (Rank 2)

Let matrix A have rows [1, 2, 3], [4, 5, 6], and [7, 8, 9]. Subtract 4 times row 1 from row 2 to produce [0, −3, −6]. Subtract 7 times row 1 from row 3 to produce [0, −6, −12]. Finally, subtract 2 times the updated row 2 from row 3 to produce [0, 0, 0]. The RREF has two non-zero pivot rows and one zero row, so rank(A) = 2. The determinant of this matrix equals 0, confirming it is singular. Row 3 equals twice row 2 minus row 1, a textbook case of linear dependence among rows.

Worked Example: Full Rank Matrix (Rank 3)

Consider matrix B with rows [2, 1, 0], [0, 3, 1], and [1, 0, 4]. Gaussian elimination produces three pivot rows with no zero row in the RREF. Therefore rank(B) = 3. The determinant of B equals 25, confirming full rank. B is invertible, and the linear system Bx = b has exactly one solution for any vector b. Any 3×3 matrix with a non-zero determinant achieves this maximum rank of 3.

The Rank-Nullity Theorem

Rank connects to the null space of A through the Rank-Nullity Theorem: for any m×n matrix, rank(A) + nullity(A) = n. For a 3×3 matrix this becomes rank(A) + nullity(A) = 3. As detailed in the Math 240 Linear Systems lecture notes (University of Pennsylvania), nullity counts the number of free variables when solving Ax = 0. If rank(A) = 2, nullity = 1, meaning one free variable exists and the solution to Ax = b (when consistent) forms a one-dimensional affine subspace. The UC Davis Linear Algebra textbook provides complete proofs connecting pivot count to solution uniqueness across all cases.

Practical Applications of Matrix Rank

Linear systems: rank(A) = 3 guarantees a unique solution to Ax = b for any right-hand side b; rank below 3 produces either infinitely many solutions or no solution — essential knowledge for engineering and physics problem-solving.
Computer graphics and 3D transformations: A 3×3 transformation matrix of rank 2 collapses 3D space onto a 2D plane; rank 1 collapses space to a line; full rank 3 preserves the full dimensionality of the transformation.
Machine learning and data compression: Low-rank matrix approximations underpin Principal Component Analysis, recommendation engines, and image compression by retaining only the dominant structural components of large data matrices.
Control systems engineering: The rank of a controllability matrix determines whether a dynamic system can be driven to any desired state — a prerequisite for designing effective feedback controllers in robotics and aerospace applications.

Reference

Frequently asked questions

What does a matrix rank of 2 mean for a 3×3 matrix?

A rank of 2 means exactly two rows — and exactly two columns — of the matrix are linearly independent. The third row is expressible as a linear combination of the other two. This makes the matrix singular, so its determinant equals zero. The linear system Ax = b then has either infinitely many solutions or no solution at all, depending on whether the vector b lies in the column space of A.

How do you find the rank of a 3×3 matrix by hand?

Apply Gaussian elimination using three types of row operations: swap two rows, multiply a row by a non-zero scalar, or add a multiple of one row to another. Continue until the matrix is in Row Echelon Form or RREF, then count the number of rows containing at least one non-zero entry. That count is the rank. For a 3×3 matrix, the result will be 0, 1, 2, or 3. A matrix rank calculator automates each elimination step instantly.

Can a 3×3 matrix have a rank of 0?

Yes, but only when every one of the nine entries equals zero. The zero matrix has an empty column space with dimension zero, giving rank(A) = 0. Any matrix containing at least one non-zero entry has rank of at least 1. In practice, a matrix rank calculator returning 0 indicates that all nine entries from a₁₁ through a₃₃ were entered as zero, which is the only scenario producing this result.

What is the difference between matrix rank and the determinant?

The determinant is a single scalar value that equals zero for any singular 3×3 matrix, while rank is an integer from 0 to 3 measuring the dimension of the column space. A non-zero determinant guarantees rank = 3. A zero determinant confirms rank is less than 3, but it cannot distinguish between rank 1 and rank 2 — both produce a zero determinant. Matrix rank therefore provides strictly more structural information than the determinant alone.

How does the rank of a 3×3 matrix affect solutions to a linear system?

For the linear system Ax = b with a 3×3 coefficient matrix A, the rank determines the solution structure. If rank(A) = 3, the system has exactly one unique solution for any vector b. If rank(A) = 2, the null space has dimension 1, yielding either infinitely many solutions or no solution. The Rank-Nullity Theorem states nullity = 3 minus rank(A), which equals the number of free variables — directly counting the degrees of freedom in the solution set.

What is the maximum possible rank of a 3×3 matrix?

The maximum rank of a 3×3 matrix is 3, achieved when all three rows are linearly independent and together span all of three-dimensional space ℝ³. This condition is called full rank. A full-rank 3×3 matrix is always invertible, carries a non-zero determinant, and satisfies the property that Ax = 0 only when x is the zero vector. Randomly generated 3×3 matrices with real-valued entries are full rank with probability 1, making rank-deficient matrices a special case in practice.