Background Stigma among healthcare providers is a barrier to the effective delivery of mental health services in primary care. Few studies have been conducted in primary care settings comparing the attitudes of healthcare providers and experiences of people with mental illness who are service users in those facilities. Such research is necessary across diverse global settings to characterize stigma and inform effective stigma reduction. Methods Qualitative research was conducted on mental illness stigma in primary care settings in one low-income country (Nepal), two lower-middle income countries (India, Tunisia), one uppermiddle- income country (Lebanon), and three high-income countries (Czech Republic, Hungary, Italy). Qualitative interviews were conducted with 248 participants: 64 primary care providers, 11 primary care facility managers, 111 people with mental illness, and 60 family members of people with mental illness. Data were analyzed using framework analysis. Results Primary care providers endorsed some willingness to help persons with mental illness but reported not having appropriate training and supervision to deliver mental healthcare. They expressed that people with mental illness are aggressive and unpredictable. Some reported that mental illness is incurable, and mental healthcare is burdensome and leads to burnout. They preferred mental healthcare to be delivered by specialists. Service users did not report high levels of discrimination from primary care providers; however, they had limited expectations of support from primary care providers. Service users reported internalized stigma and discrimination from family and community members. Providers and service users reported unreliable psychiatric medication supply and lack of facilities for confidential consultations. Limitations of the study include conducting qualitative interviews in clinical settings and reliance on clinician-researchers in some sites to conduct interviews, which potentially biases respondents to present attitudes and experiences about primary care services in a positive manner. Conclusions Primary care providers' willingness to interact with people with mental illness and receive more training presents an opportunity to address stigmatizing beliefs and stereotypes. This study also raises important methodological questions about the most appropriate strategies to accurately understand attitudes and experiences of people with mental illness. Recommendations are provided for future qualitative research about stigma, such as qualitative interviewing by non-clinical personnel, involving non-clinical staff for recruitment of participants, conducting interviews in non-clinical settings, and partnering with people with mental illness to facilitate qualitative data collection and analysis.