We report the development and psychometric evaluation of short forms of the Oral Health Impact Profile German version (OHIP-G) - an instrument to assess oral health-related quality of life (OHRQoL). A five-item short form was developed using best subset regression in 2050 subjects from a national survey. Two 14-item versions were derived from English-language short forms and a 21-item version from previous factor analytic work. A second sample from the general population (n = 163) and a sample of clinical patients with temporomandibular disorders (TMD; n = 175) were used to investigate validity and internal consistency. Test-retest reliability was evaluated in 30 prosthodontic patients before treatment. Responsiveness was assessed in 67 patients treated for their TMD pain. Associations between short form summary scores and self-report of oral health and four oral disorders in the general population and in TMD patients were interpreted as support for convergent/groups validity. The instruments' responsiveness (effect measures of 0.55-0.98), test-retest reliability (intraclass correlation coefficients: 0.72-0.87), and internal consistency (Cronbach's alpha: 0.65-0.92) were sufficient. Sufficient discriminative and evaluative psychometric properties of short forms of the OHIP-G make the instruments suitable to assess OHRQoL in cross-sectional as well as longitudinal studies.