Polyether-ether-ketone (PEEK) is a potential alternative to metal alloys for use in the orthopedic implants; however, an in vivo study in an appropriate animal model is essential and has not yet been initiated. The aim of the present study was to gather some preliminary evidence regarding the performance and safety of a cemented PEEK-based knee replacement device in a goat model. Fifteen adult goats were randomly divided into two groups; the control group (n = 5), which received a sham operation, and the experimental group (n = 10), which received a total knee arthroplasty with a PEEK-based knee replacement device. The animals were sacrificed at 12 (control n = 5; experimental n = 5) or 24 weeks (experimental n = 5). Blood parameter measurements and radiographs of the knee joints were obtained. The synovium and main organs were removed and histologically assessed. The knee joints with the prosthesis were analyzed via micro-computed tomography and laser confocal microscopy. There was no occurrence of implant fracture or prosthesis sinking during the 24 weeks of radiological observations, except for one case of prosthesis dislocation at 4 weeks. There was a 6% decrease in femoral bone density (BD) at 12 weeks, but no further decrease by 24 weeks. No changes in BD were observed in the tibial ends. The bulk implant was biocompatible in terms of histological analysis of the local synovium and organs. There were no optical scratches on the surface of the retrieved components; the femoral component surface was rougher, while the tibial insert was smoother after 24 weeks. The novel PEEK-based knee replacement device in a goat model was feasible and safe; however, prior to use in humans, further studies concerning PEEK high load-bearing implant designs should be carried out to expand on our results.