Purpose: To compare the performance of microdose flare-up (MF) and flexible-multidose gonadotropin-releasing hormone (GnRH) antagonist protocols in poor responder patients who underwent intracytoplasmic sperm injection (ICSI). Materials and Methods: One hundred and 12 consecutive patients (217 cycles) suspected to have poor ovarian response were enrolled. Group 1 (MF GnRH agonist group) constituted 64 patients (135 cycles) who underwent MF GnRH agonist protocol. Group 2 (flexible-multidose GnRH antagonist group) constituted 48 patients (82 cycles) who underwent flexible-multidose GnRH antagonist protocol. Results: The duration of stimulation (d) (11.5 ± 2.1 vs. 10.4 ± 2.7, p < 0.01) and the total dose of gonadotropin used (IU) (5,892.9 ± 1,725.7 vs. 4,367.5 ± 1,582.1, p <0.05) were significantly lower in Group 2 when compared to Group 1. The numbers of retrieved oocyte-cumulus complexes (4.5 ± 3.6 vs. 5.9 ± 4.9, p < 0.05), metaphase II oocytes (3.6 ± 3.1 vs. 4.9 ± 4.2, p < 0.05), two pronucleated oocytes (2.6 ± 2.3 vs. 4.0 ± 3.4, p < 0.05), the number of available embryos at day 3 (2.6 ± 2.2 vs. 4.2 ± 3.2, p < 0.05) and the rate of embryos with ≥ seven blastomeres and < 10% fragmentation at day 3 (35.9% vs. 65.1%, p < 0.05) were significantly lower in Group 1 when compared to Group 2. The number of embryos transferred (2.2 ± 1.3 vs. 2.4 ± 0.9), the clinical pregnancy per embryo transfer (16.3% vs. 25.8%), and the implantation rate (8.6% vs. 12.2%) were comparable between groups. Conclusions: Although the flexible-multidose GnRH antagonist protocol produced better oocyte and embryo parameters, the clinical pregnancy rate and the implantation rates were comparable between the flexible-multidose GnRH antagonist and MF protocols in poor responder patients.