This study aimed to characterize the metabolic composition of four types of commercially available chicken breeds [village chicken, colored broiler (Hubbard), broiler (Cobb), and spent layers (Dekalb)] by
1
H NMR coupling and discriminate them using multivariate analysis. Five chickens were collected for each chicken breed based on the marketing age from the respective commercial farms. The orthogonal partial least squares discriminant analysis (OPLS‐DA) results showed an obvious separation of local village chickens from the other breeds based on the metabolites present in their serum and meat (pectoralis major). The cumulative values of
Q
2
,
R
2
X
, and
R
2
Y
of the OPLS‐DA model for chicken serum were 0.722, 0.877, and 0.841. For the pectoralis major muscle, the cumulative values of
Q
2
,
R
2
X
, and
R
2
Y
of the OPLS‐DA model were reported as 0.684, 0.781, and 0.786, respectively. The quality of both OPLS‐DA models was accepted by the cumulative values of
Q
2
≥ 0.5 and
R
2
≥ 0.65. The
1
H NMR result with multivariate analysis has successfully distinguished local village chicken from the other three commercial chicken breeds based on serum and pectoralis major muscle. Nonetheless, colored broiler (Hubbard) was not distinguished from broiler (Cobb) and spent layers (Dekalb) in serum and pectoralis major, respectively. The OPLS‐DA assessment in this study identified 19 and 15 potential metabolites for discriminating different chicken breeds in serum and pectoralis major muscle, respectively. Some of the prominent metabolites identified include amino acids (betaine, glycine, glutamine, guanidoacetate, phenylalanine, and valine), nucleotides (IMP and NAD+), organic acids (lactate, malate, and succinate), peptide (anserine), and sugar alcohol (myo‐inositol).