Is the formula different in 2D and 3D?

No. 2D is the case where z components are zero for both vectors.

Why must vectors be nonzero?

Division by |a| or |b| is undefined when a magnitude is zero.

What if cos θ is slightly above 1?

Round-off error. Clamp to 1 before arccos.

Can I use the law of cosines instead?

For triangles built from vector lengths and the angle between them, yes. The dot form is the direct vector route.

Where does θ = arccos((a·b)/(|a||b|)) fail?

When either vector has zero length or when components are not in the same coordinate frame.

Angle Between Two Vectors Formula

Every angle calculation on this site reduces to one cosine identity. Here is the full formula stack: component dot product, magnitudes, division, and inverse cosine.

Open the calculator

Quick answer

θ = arccos( (a · b) / (|a| |b|) ) with a · b = aₓbₓ + aᵧbᵧ + a_z b_z and |a| = √(aₓ² + aᵧ² + a_z²).

Formula

cos θ = (a · b) / (|a| |b|)
θ = arccos(cos θ)

Introduction

The cosine rule for vectors is the bridge between algebra and geometry. The Angle Between Two Vectors Calculator evaluates the same expression after you supply components or point pairs.

Start with what is the angle between two vectors if you want the geometric definition first.

What each part of the formula does

The dot product a · b measures alignment along shared axes. Magnitudes |a| and |b| measure lengths. Dividing removes scale so only direction remains inside the cosine.

arccos recovers the angle in radians. Multiply by 180/π when readers expect degrees.

The cosine rule is the vector version of a fact you already know from triangles: the dot encodes how much two directions overlap when lengths are accounted for.

Each symbol has a job: components feed the dot, magnitudes normalize, arccos inverts cosine. Skip any step and the angle loses meaning.

Full formula block

a · b = aₓbₓ + aᵧbᵧ + a_z b_z
|a| = √(aₓ² + aᵧ² + a_z²)
cos θ = (a · b) / (|a| |b|)
θ = arccos(cos θ)

Clamp cos θ to [-1, 1] before arccos when floating-point noise appears. Zero magnitudes make the expression undefined.

For intuition behind the dot product term, read dot product and vector angle explained.

Using the formula

Compute a · b from components. Multiply matching components and add. In 3D include z with z; in 2D confirm both z terms are truly zero.
Compute |a| and |b|. Square, sum, square root for each vector separately. A calculator memory line for each magnitude reduces typos.
Form the cosine ratio. Divide dot by the product of magnitudes. Write the fraction before you decimalize; reviewers like seeing the exact ratio.
Apply arccos and label units. Radians by default in many libraries; degrees for many textbooks.
Check special cosine values. 0, 1, -1, 0.5, and √2/2 correspond to familiar angles and catch many mistakes instantly.

Formula in numbers

For a = (3, 0) and b = (3, 4): a·b = 9, |a| = 3, |b| = 5, cos θ = 0.6, θ ≈ 53.13°.

More patterns appear in angle between two vectors examples.

Perpendicular check: a = (2, -1) and b = (1, 2) give dot 0 with nonzero lengths, so cos θ = 0 and θ = 90° without needing a calculator.

Frequently asked questions

Is the formula different in 2D and 3D?: No. 2D is the case where z components are zero for both vectors.
Why must vectors be nonzero?: Division by |a| or |b| is undefined when a magnitude is zero.
What if cos θ is slightly above 1?: Round-off error. Clamp to 1 before arccos.
Can I use the law of cosines instead?: For triangles built from vector lengths and the angle between them, yes. The dot form is the direct vector route.
Where does θ = arccos((a·b)/(|a||b|)) fail?: When either vector has zero length or when components are not in the same coordinate frame.

Conclusion

Memorize the cosine ratio, not a forest of variants. Dot, magnitudes, divide, arccos, then unit label.

Use the step-by-step article when you want a checklist from raw inputs to θ.

Carry the formula on one index card: dot on the front, magnitudes on the back. Most exam errors are bookkeeping, not calculus.