Rhizoma peanut (RP, Arachis glabrata) is an alternative to alfalfa (ALF, Medicago sativa L.) with less nonstructural carbohydrates (NSC) and crude protein. Yet, health variables in response to feeding RP have not been evaluated. This study investigated fecal and blood health markers in horses fed forage-only diets of ‘Florigraze’ RP, ‘Legendary XHD’ alfalfa, or ‘Coastal’ bermudagrass (BMG, Cynodon dactylon L.) hays. Six mature, healthy Quarter Horse geldings (593 kg ± SD 40) were fed the 3 hays in a replicated 3 × 3 Latin square with 3 21-d periods each separated by a 4-d washout. Hays were fed at 2% BW with ad libitum water and mineral supplement to meet maintenance requirements. Crude protein was 22, 16, and 8% and NSC was 11, 10, and 7.8% in ALF, RP, and BMG, respectively. On d 19 fecal and blood samples were collected immediately before (0 h) and 2, 4, and 8 h after feeding. Data were compared using mixed model ANOVA with repeated measures. Fecal dry matter, lactate, and pH averaged (±SEM) 19 ± 0.73%, 2.08 ± 0.172 mM, and 7.3 ± 0.05, respectively and did not differ by hay type. Fecal ammonia-nitrogen was greatest (P = 0.02) when fed ALF, but similar between RP and BMG. Fecal total short-chain fatty acids (SCFA) did not differ among hays (P = 0.07). Expressed as molar proportion, hay type did not alter acetate proportion (P = 0.12) but affected proportions of propionate (P = 0.02), butyrate (P = 0.04), andbranched-chain (BC) SCFA (isobutyrate + isovalerate + 2-methylbutyrate). Propionate proportion was greater (P = 0.02) when fed ALF than RP, but similar to both legumes when fed BMG. Butyrate proportion was greater (P = 0.04) for BMG-fed horses than RP-fed horses, while ALF was similar to both. Molar proportion of BC-SCFA was similar for both legume hays and least (P = 0.02) for BMG. Except for dry matter and lactate, time post feeding did not affect any of the fecal variables. Fecal dry matter decreased (P < 0.01) at 2 h and 4 h post feeding and had returned to pre-feeding values at 8 h post feeding. Fecal lactate decreased (P < 0.01) only at 8 h post feeding. Alfalfa resulted in glycemic (P = 0.01) and insulinemic (P = 0.01) responses greater than BMG, whereas the responses to RP were similar to those observed for ALF and BMG. Serum glucose was affected by time (P = 0.01) where it was elevated at 4 h post feeding. Insulin response was not affected by time post feeding (P = 0.70). Rhizoma peanut resulted in plasma urea nitrogen (PUN) lesser than ALF and similar to BMG. Levels of PUN decreased (P < 0.01) 8 h post feeding. Feeding rhizoma peanut provides more favorable metabolic responses than alfalfa, while still providing a high-quality source of nutrition.