There are many surgical techniques (packing, Pringle maneuver, etc.) and hemostatic agents to manage hepatic bleeding in trauma surgery. This study compares the effectiveness of two different types of hemostatic agents, one is an active flowable hemostat and the other is a passive hemostat made of modified absorbable polymers [MAP]. Both surgical technique and hemostatic agents can be used together as a means of controlling bleeding. We have hypothesized that a single hemostatic agent might be as effective as a unique hemostatic surgical technique. Twenty swine were prospectively randomized to receive either active Flowable (Floseal) or passive MAP powder (PerClot) hemostatic agents. We used a novel severe liver injury model that caused exsanguinating hemorrhage. The main outcome measure was total blood loss volume. The total volume of blood loss, from hepatic injury to minute 120, was significantly lower in the Flowable group (407.5 cm3; IqR: 195.0–805.0 cm3) compared to MAP group (1107.5 cm3; IqR: 822.5 to 1544.5 cm3) (Hodges–Lehmann median difference: − 645.0 cm3; 95% CI: − 1144.0 to − 280.0 cm3; p = 0.0087). The rate of blood loss was significantly lower in the flowable group compared with the MAP group as measured from time of injury to minutes 3, 9, 12, and 120 (except for 6 min). The mean arterial pressure gradually recovered in the flowable group by 24 h, whereas in the MAP group, the mean arterial pressure was consistently stayed below baseline values. Kaplan–Meier survival analysis indicated similar rates of death between study groups (Logrank test p = 0.3395). Both the flowable and the MAP hemostatic agents were able to effectively control surgical bleeding in a novel severe liver injury model, however, the flowable gelatin–thrombin agent provided quicker and better bleed control.