The present study chiefly aimed to compare two forms of dynamic assessment and standard assessment of EFL learners’ listening comprehension. 59 Iranian EFL learners were randomly assigned to three test administration groups and assessed on listening for main and supporting information in listening tasks. The first model of dynamic assessment was a form of scaffolding in which for any incorrect answer graded prompts were progressively presented to learners until they answered correctly. The second model of dynamic assessment was direct instruction of listening tasks. Learners in standard assessment group, however, completed the task independently without mediation. The results revealed statistically significant listening improvement in favor of dynamic-supported and dynamic-instructed assessment groups rather than standard assessment one. The findings of the study indicate that dynamic assessment can gain better insights into learners’ level of comprehension and their potential for future development and provide better learning effectiveness than those in statistic assessments.