This study investigated how input frequency (i.e., type frequency and token frequency) and proficiency levels enhanced the perception of English nominal suffixes by first language (L1) Thai learners. Based on the Usage-based Account (Tomasello, 2003), it was hypothesized that input frequency, i.e., token frequency (frequency of derived forms containing the particular suffix) and type frequency (suffix frequency), facilitates SLA. A Grammaticality Judgement Test (GJT) was administered to 60 L1 Thai learners at the intermediate and the advanced proficiency levels, 30 per group. The four frequency conditions were as follows: Condition 1 (HH) -- high type and high token frequency (e.g., 'alteration'); Condition 2 (HL) -- high type and low token frequency (e.g., 'chemist'); Condition 3 (LH) -- low type and high token frequency (e.g., 'dependence'); and Condition 4 (LL) -- low type and low token frequency (e.g., 'partnership'). The results showed significant effects of input frequency and proficiency levels. However, the interaction between them was not significant. The study also revealed that the intermediate group perceived Condition 3 (LH) most accurately, whereas Condition 1 (HH) was perceived most accurately by the advanced group. The findings supported the Usage-based Account, indicating that input frequency influenced SLA of English nominal suffixes, with token frequency having a greater impact than type frequency.