There is a controversy about potentially positive or negative effects of caffeine consumption on onset and disease progression of neurodegenerative diseases such as Huntington’s Disease (HD). On the molecular level, the psychoactive drug caffeine targets in particular adenosine receptors (AR) as a nonselective antagonist. The aim of this study was to evaluate clinical effects of caffeine consumption in patients suffering from premanifest and motor-manifest HD. Data of the global observational study ENROLL-HD were used, in order to analyze the course of HD regarding symptoms onset, motor, functional, cognitive and psychiatric parameters, using cross-sectional and longitudinal data of up to three years. We split premanifest and manifest participants into two subgroups: consumers of >3 cups of caffeine (coffee, cola or black tea) per day (>375 mL) vs. subjects without caffeine consumption. Data were analyzed using ANCOVA-analyses for cross-sectional and repeated measures analysis of variance for longitudinal parameters in IBM SPSS Statistics V.28. Within n = 21,045 participants, we identified n = 1901 premanifest and n = 4072 manifest HD patients consuming >3 cups of caffeine/day vs. n = 841 premanifest and n = 2243 manifest subjects without consumption. Manifest HD patients consuming >3 cups exhibited a significantly better performance in a series of neuropsychological tests. They also showed at the median a later onset of symptoms (all p < 0.001), and, during follow-up, less motor, functional and cognitive impairments in the majority of tests (all p < 0.050). In contrast, there were no beneficial caffeine-related effects on neuropsychological performance in premanifest HD mutation carriers. They showed even worse cognitive performances in stroop color naming (SCNT) and stroop color reading (SWRT) tests (all p < 0.050) and revealed more anxiety, depression and irritability subscores in comparison to premanifest participants without caffeine consumption. Similarly, higher self-reported anxiety and irritability were observed in genotype negative/control group high dose caffeine drinkers, associated with a slightly better performance in some cognitive tasks (all p < 0.050). The analysis of the impact of caffeine consumption in the largest real-world cohort of HD mutation carriers revealed beneficial effects on neuropsychological performance as well as manifestation and course of disease in manifest HD patients while premanifest HD mutation carrier showed no neuropsychological improvements, but worse cognitive performances in some tasks and exhibited more severe signs of psychiatric impairment. Our data point to state-related psychomotor-stimulant effects of caffeine in HD that might be related to regulatory effects at cerebral adenosine receptors. Further studies are required to validate findings, exclude potential other unknown biasing factors such as physical activity, pharmacological interventions, gender differences or chronic habitual influences and test for dosage related effects.