Coffee intake has been inversely related to the incidence of liver diseases, although there are controversies on whether these beneficial effects on human health are because of caffeine or other specific components in this popular beverage. Thus, this study evaluated the protective effects of coffee or caffeine intake on liver injury induced by repeated thioacetamide (TAA) administration in male Wistar rats. Rats were randomized into five groups: one untreated group (G1) and four groups (G2–G5) treated with the hepatotoxicant TAA (200�mg/kg b.w., i.p.) twice a week for 8�weeks. Concomitantly, rats received tap water (G1 and G2), conventional coffee (G3), decaffeinated coffee (G4) or 0.1% caffeine (G5). After 8�weeks of treatment, rats were killed and blood and liver samples were collected. Conventional and decaffeinated coffee and caffeine intake significantly reduced serum levels of alanine aminotransferase (ALT) (p�<�0.001) and oxidized glutathione (p�<�0.05), fibrosis/inflammation scores (p�<�0.001), collagen volume fraction (p�<�0.01) and transforming growth factor β‐1 (TGF‐β1) protein expression (p�≤�0.001) in the liver from TAA‐treated groups. In addition, conventional coffee and caffeine intake significantly reduced proliferating cellular nuclear antigen (PCNA) S‐phase indexes (p�<�0.001), but only conventional coffee reduced cleaved caspase‐3 indexes (p�<�0.001), active metalloproteinase 2 (p�≤�0.004) and the number of glutathione S‐transferase placental form (GST‐P)‐positive preneoplastic lesions (p�<�0.05) in the liver from TAA‐treated groups. In conclusion, conventional coffee and 0.1% caffeine intake presented better beneficial effects than decaffeinated coffee against liver injury induced by TAA in male Wistar rats.