Equine producers frequently utilize legumes, such as alfalfa ( Medicago sativa ) and perennial peanut ( Arachis glabrata ), as forages in their feeding programs. Due to challenges related to production and shipping in the Southeast, producers may be looking for other viable legume options to reduce feeding costs. Soybean ( Glycine max L.) may prove to be a useful option. The current study evaluated 3 forage soybean varieties (V), ‘Tower of Leaves’ (T), ‘Stonewall’ (S), and ‘Laredo’ (L), and 2 grain varieties, ‘AG64 × 8’ (A) and ‘AG79 × 9’ (X), during the 2019–2020 growing seasons for forage yield and nutritive value. Row spacing (RS; 36 and 71 cm), harvest height (HH; 10, 15, and 20 cm), and location [E.V. Smith Research and Extension Center (EVS) and Wiregrass Research and Extension Center (WG)] were also evaluated. Plots were harvested using a small plot harvester. A subsample was dried in a 55°C forced air oven and then ground to 1 mm. Crude protein (CP), neutral detergent fiber (NDF), and acid detergent fiber (ADF) was determined using NIRS analysis. Wet chemistry was performed on 10% of samples for verification. In vitro dry matter digestibility (IVDMD) was determined using batch culture inoculated with an equine fecal matter (10%) and buffer solution (90%) mixture. This study was a randomized, complete block design (n = 4) and stats were analyzed using PROC Glimmix of SAS version 9.4 (SAS Inst., Cary, NC). There was no effect of RS on any parameter evaluated ( P > 0.17). All dependent variables were affected by HH ( P < 0.01), V ( P < 0.01), and location ( P < 0.05). Yield was greatest in S and T ( P < 0.02; 2074 and 1961 kg ha −1 , respectively), while yield was least in L ( P < 0.01; 1340 kg ha −1 ). Neutral detergent fiber and ADF were less in S than L ( P < 0.01; 46 and 37 vs 47 and 38%, respectively), and A ( P < 0.02; 46 and 37 vs 47 and 38%, respectively), while X was also less than L ( P < 0.01; 46 and 37 vs 47 and 38%, respectively) and T ( P < 0.03; 46 and 37 vs 47 and 38%, respectively). Crude protein was greatest in A and X ( P < 0.05; 16.9 and 16.4%, respectively), while it was least in S and T ( P < 0.05; 15.8 and 15.7%, respectively). There was no effect ( P = 0.20) of soybean variety on IVDMD. The EVS location had greater forage yield ( P < 0.0001), CP ( P < 0.0001), and NDF ( P < 0.05), but lesser ADF ( P < 0.0001) than WG. The cultivar S showed the greatest forage yield and the least NDF and ADF, while the CP seen in S was less than other varieties; however, S was only 1% lower than A, which had the greatest CP. This indicates that S has the greatest promise as a forage option and should be further tested in mixtures with other forage species.