"Coprime" and "Prime" Sound the Same — They Describe Completely Different Things

You read a problem that says "let a and b be coprime," you pattern-match on the sound of the word, and your brain quietly substitutes "let a and b be prime." Then the whole solution goes sideways. This is one of the most common vocabulary traps in early number theory, and it is worth spending a few minutes killing the confusion permanently.

Coprime (also called relatively prime or mutually prime) is a property of two numbers considered together. Prime is a property of one number on its own. They are different kinds of statements about different kinds of objects. Mistaking one for the other is like mistaking "tall" (a property of a person) for "taller than" (a relationship between two people).

The definitions side by side

Prime

An integer p > 1 is prime if its only positive divisors are 1 and p itself.

Coprime

Two integers a and b are coprime if \gcd(a, b) = 1 — that is, they share no positive common divisor other than 1.

Read the first definition: the input is a single number. You either are or are not prime.

Read the second: the input is a pair of numbers. The question "is 8 coprime?" is not even well-formed — coprime to what?

The sharpest example

Take a = 8 and b = 9.

8 = 2^3. Is 8 prime? No — it has divisors 1, 2, 4, 8. It is composite.
9 = 3^2. Is 9 prime? No — it has divisors 1, 3, 9. It is composite.
Are 8 and 9 coprime? Yes — they share no prime factor (one is all 2s, the other all 3s), so \gcd(8, 9) = 1.

Two composite numbers, coprime to each other. This example alone proves the two concepts live in different universes.

Why the names sound so similar

The word coprime was chosen because of a useful fact: if a and b are coprime, they behave like primes toward each other in the arithmetic identity

\gcd(a, b) \cdot \operatorname{lcm}(a, b) = a \cdot b.

When \gcd(a, b) = 1, this collapses to \operatorname{lcm}(a, b) = a \cdot b. So coprime numbers multiply together cleanly, with no overlap to subtract out — exactly the way two distinct primes would. The prefix "co-" is the Latin root meaning "together" or "jointly"; coprime means "prime with respect to each other," not "each one individually prime."

A quick Venn picture

Imagine the prime factors of a number as marbles in a bag. The prime factors of 8 are three marbles all coloured "2." The prime factors of 9 are two marbles all coloured "3."

The left bag holds the primes of $8$ (three $2$s). The right holds the primes of $9$ (two $3$s). The overlap is empty — that is coprimality. Whether the bags themselves contain a single marble (prime) or many (composite) is a completely separate question.

Two numbers are coprime exactly when their bags share no colour. Two numbers are prime when each bag contains exactly one marble.

Try the definitions live

a: b:

Drag $a$ and $b$ between $2$ and $60$. The visualiser factorises each number live, computes $\gcd(a, b)$ by the Euclidean algorithm, and tells you which of the two properties (prime, coprime) each side has. Try $(8, 9)$, $(12, 25)$, $(7, 11)$, $(12, 18)$ — all four combinations of the table appear.

The logical relationships

There are four combinations of "each number prime" and "pair coprime." All four can happen.

Each number prime?	Pair coprime?	Example
Yes	Yes	\{2, 3\}, \{7, 11\} — any two distinct primes
Yes	No	\{5, 5\} — same prime twice, \gcd = 5
No	Yes	\{8, 9\}, \{4, 25\} — composites with disjoint prime supports
No	No	\{12, 18\}, \{6, 10\} — composites that share a prime

The only implication that holds is the first row's shape: two distinct primes are always coprime. That one is genuine. But notice it says "two distinct primes" — the coprime conclusion is about the pair, not about either number individually.

Why the distinction matters in practice

Many theorems in number theory have the hypothesis "a and b are coprime" — and misreading it as "a and b are prime" loses most of the theorem's applications.

Euler's totient multiplicativity. \varphi(ab) = \varphi(a)\varphi(b) when a and b are coprime. Since \varphi(8) = 4 and \varphi(9) = 6 and \varphi(72) = 24 = 4 \cdot 6, the formula applies to 8 and 9 — both composite. If the hypothesis were "both prime," the theorem would cover only a handful of cases.
Chinese Remainder Theorem. A system x \equiv r_1 \pmod{m_1}, x \equiv r_2 \pmod{m_2} has a unique solution mod m_1 m_2 when m_1 and m_2 are coprime. Most useful applications take m_1 = 4 and m_2 = 9 or similar — composite but coprime.
Fraction in lowest terms. A fraction p/q is in lowest terms when \gcd(p, q) = 1 — that is, when p and q are coprime. p and q are almost never both prime. 7/12 is in lowest terms, and 12 is composite.

In all three, replacing "coprime" with "prime" would make the theorem nearly useless.

The mental model that fixes this forever

When you hear the word "prime," think one number, one question: does it break into smaller factors?

When you hear "coprime," think two numbers, one question: do they share any prime factor?

The two questions have no logical connection to each other. Keep them in separate drawers of your mind, and the confusion stops.