Previous research shows that children with autism spectrum disorder (ASD) may experience several difficulties-including those related to eating-and this area of research needs to be explored further. This study had two main objectives: (1) comparison of the clinical (autism spectrum disorder) and non-clinical sample of children in terms of avoidant/restrictive food disorder, food neophobia, other eating-related behaviours and feeding practices; (2) assessment of selected predictors of food neophobia. The final sample included 54 children and parents from the clinical sample (ASD) and 51 from the non-clinical sample. Parents completed: the autism spectrum rating scales (ASRS), the eating disorders in youth questionnaire (EDY-Q), the children food neophobia scale (CFNS), the child eating behaviour questionnaire (CEBQ), the child feeding questionnaire (CFQ), and a socio-demographic survey. Our analysis allowed us to partially confirm the first hypothesis since the clinical sample (vs. the non-clinical group) had significantly higher scores in such variables as (a) avoidant/restrictive food disorder (ARFID), (b) food neophobia, (c) other eating-related behaviours: emotional under-eating, desire to drink, food fussiness, (d) feeding practices: pressure to eat. Moreover, our analysis of predictors of food neophobia in the clinical and non-clinical samples also allowed us to partially confirm the second hypothesis because it turned out that only in the clinical sample (vs. the non-clinical sample) were the predictors significantly associated with food neophobia, but only two of them (food fussiness and selective eating). In conclusion, our study showed that children with ASD (compared to children without this diagnosis) experience increased difficulties in eating behaviours, and their parents present a stronger intensity of the feeding practice based on pressure to eat. This study showed that feeding problems for children in the ASD sample are a significant problem, and it is still worth exploring this area in research.