What actually makes something a 'peptide' vs a protein? Where's the line?

It's genuinely a bit of a convention rather than a hard law of nature. The commonly cited line is around 50 amino acid residues, below which people tend to say 'peptide' and above which 'protein,' but there's no committee enforcing it. Function and folding matter more than the raw count.

Insulin is the classic edge case you noticed. It's small (two chains, ~51 residues total), so some call it a peptide hormone and others a small protein. Both are defensible. The takeaway: don't over-index on the exact number, it's a spectrum.

Adding one nuance: proteins usually have meaningful secondary/tertiary structure (they fold into a shape that does the work), while many peptides are short enough that they don't fold into a stable 3D structure on their own. That functional/structural distinction is arguably more useful than counting residues.

This is exactly what I needed, thank you both. The folding framing clicked for me way more than the amino-acid count did. Off to read the glossary entry on primary vs tertiary structure.