Add, subtract, multiply, transpose, and find the determinant or inverse of a 2x2 or 3x3 matrix. No data ever leaves your device.
A matrix inverse gets used to solve systems of linear equations, and by hand it means a page of cofactor arithmetic where one sign error anywhere invalidates the whole answer. Matrix Calculator runs the same cofactor-expansion method for 2×2 and 3×3 matrices instantly, alongside addition, subtraction, multiplication and transpose.
Reach for it to check homework by hand, verify a linear algebra step in a larger calculation, or get a quick determinant without opening a full computer algebra system. Switch between 2×2 and 3×3 at any time; changing size resets both matrices to zero.
The 2×2 determinant is ad − bc; the 3×3 case expands along the first row into three 2×2 determinants with alternating signs. The inverse is the adjugate matrix (the cofactor matrix, transposed) divided by the determinant — the standard method taught in a first linear algebra course. Every formula here was checked during development against NumPy’s linalg.det and linalg.inv on two test matrices and matched exactly, including a 3×3 case with a determinant of exactly 1.
A matrix with a determinant of 0 has no inverse by definition — [[1,2],[2,4]] is the simplest example, since its second row is just the first row doubled. Clicking A⁻¹ on a matrix like that shows “Matrix is singular (determinant = 0) — it has no inverse.” rather than dividing by zero silently. Results are rounded to 4 decimal places for display, so an inverse that’s mathematically exact may show a value like 0.3333 instead of the fraction it actually is.
Size
Operation
Matrix A
Matrix B
Result
3×3 determinant uses cofactor expansion along the first row, and the 3×3 inverse is the cofactor matrix’s transpose (the adjugate) divided by the determinant — the same method taught in a first linear algebra course, verified during development against NumPy’s own linalg.det and linalg.inv. Switching size resets both matrices to zero rather than trying to guess how to extend or crop your existing values.