Fungemia due to rare yeasts constitutes an emerging but poorly investigated condition. Data on risk factors, clinical features, therapy, and outcome of episodes of fungemia due to rare (non-Candida, non-Cryptococcus) yeasts were analyzed in a population-based surveillance program conducted in 29 Spanish hospitals between May 2010 and April 2011. Species identification (DNA sequencing) and antifungal susceptibility testing (EUCAST and CLSI methods) were centrally performed. Fourteen out of 767 episodes of fungemia (1.8%) were due to rare yeasts: Trichosporon asahii, Magnusiomyces capitatus (three cases each), Rhodotorula mucilaginosa, Wickerhamomyces anomalus (two cases each), and Pichia kudriavzevii, Cyberlindnera fabianii, Kodamaea ohmeri, and Lodderomyces elongisporus (one case each). Misidentification by local laboratories was observed in two isolates. Breakthrough fungemia occurred in two episodes due to M. capitatus. MIC values for echinocandins were generally high (particularly for M. capitatus, T. asahii, and R. mucilaginosa isolates [≥2 mg/l]), whereas T. asahii isolates showed MICs ≥1 mg/l to amphotericin B. Patients with fungemia due to rare yeasts were more likely to have hematological malignancies (28.6% vs. 7.8%; P-value = .021), chronic lung disease (50.0% vs. 22.3%; P-value = .023), and prior immunosuppression (57.1% vs. 22.2%; P-value = .005) compared to those with candidemia. The rate of clinical failure (persistent fungemia and/or 30-day mortality) was 46.2% and did not significantly differ from that observed in episodes of candidemia. In conclusion, non-Candida, non-Cryptococcus yeasts are uncommon causes of fungemia, with immunosuppression and chronic lung disease as predisposing factors. Outcome does not appear to be worse than that of candidemia.