Early mobilization exercises are advocated following Achilles tendon (AT) repair, but forces on the repair during passive range of motion are unknown. The extent to which these forces change with flexion of the knee is also not known. Estimated AT forces were measured using 3 models: cadaveric, uninjured subjects, and in both legs of subjects 6 weeks following unilateral AT repair. For cadaveric testing, estimated AT force was recorded using a force transducer while cycling the ankle from 10 degrees plantarflexion to maximum dorsiflexion at 3 different knee flexion angles (0 degrees , 45 degrees , and 90 degrees ). For in vivo testing, subjects were seated in an isokinetic dynamometer, and their ankles passively cycled from plantarflexion to dorsiflexion with the knee extended and flexed 50 degrees . Passive plantarflexion moment recorded by the dynamometer was converted to AT force by estimating the AT moment arm. In the cadaveric model, knee flexion reduced estimated AT forces during dorsiflexion by more than 40% (P < .036). In vivo testing showed that estimated AT force was reduced in knee flexion in healthy subjects (P < .001) and in the uninvolved leg AT repair subjects (P = .021), but not in the AT repaired leg (P = .387). Normal AT showed a marked reduction in estimated AT force with knee flexion which was not present in repaired AT. This could be because of elongation of the repair, causing more slack in the tendon that would need to be taken up before force transmission occurs. ACFAS Level of Clinical Evidence: 4.