The objective of this experiment was to determine whether faba bean could successfully be used in feeds for weaner pigs in the period 10 to 25 kg liveweight. An experiment was conducted using 48 weaner pigs (average weight 10 ± 0.42 kg) to determine the nutritive value of dehulled faba bean meal (Vicia faba cv. Fiord) in comparison with full-fat soy. Two basal feeds were formulated, the first containing no faba bean but with full-fat soy (FFS) as the protein source, while the second contained 300 g faba bean/kg feed (FB). These feeds were blended to produce a series of five feeds (T1 to T5) containing a range of faba bean contents. In addition, a choice feeding treatment was included in the experiment to determine whether pigs showed preferences for or against faba bean. The experimental feeds were: 1) 1.0 FFS; 2) 0.75 FFS + 0.25 FB; 3) 0.5 FFS + 0.5 FB; 4) 0.25 FFS + 0.75FB; 5) 1.0 FB and 6) choice between FFS and FB. Feeds were formulated to meet or exceed NRC (1998) nutrient requirements of weaner pigs. Diluting full-fat soya with dehulled faba bean meal had no effect on growth rate (ADG), feed intake (ADFI), feed conversion efficiency (FCE) or time taken to attain final weight. Male pigs had a numerically higher ADFI (1150 vs. 992 g/d) and ADG (553 vs. 539 g/d) than females, and reached the final liveweight on average 3 d before the females. When given a choice between the two basal feeds, no preference was shown for either feed. It may be concluded that a feed for weaner pigs may contain as much as 300 g dehulled faba bean/kg without causing any deleterious effects on performance as long as the quality of the faba beans is the same as that used in this trial.