Large language models (LLMs) reflect societal norms and gender stereotypes in emotion attribution. This study investigates whether emotions are gendered and whether these variations are based on societal stereotypes. Five state-of-the-art LLMs (open- and closed-source) were tested to determine if they exhibit gendered emotions. The models were prompted to adopt a gendered persona and attribute emotions to an event, such as "When I had a serious argument with a dear person." The results show that all models consistently exhibit gendered emotions, influenced by gender stereotypes. Women are associated with SADNESS, while men are associated with ANGER. These findings align with established research in psychology and gender studies. The study highlights the complex societal interplay between language, gender, and emotion. The reproduction of emotion stereotypes in LLMs allows for detailed analysis of these stereotypes but raises questions about the predictive use of LLMs for emotion applications. The study provides a quantitative analysis of over 200,000 emotion attributions across 7,000 events and two personas, spanning over 400 unique emotions. The results show that gendered stereotypes are prevalent across the models, with men more likely to be associated with ANGER and women with SADNESS. The study also includes a qualitative analysis of model explanations, revealing that models often rely on societal stereotypes when attributing emotions. The findings suggest that LLMs reflect and amplify gender stereotypes in emotion attribution, which has implications for the fairness and inclusiveness of NLP systems. The study calls for interdisciplinary collaboration to address gendered emotions in NLP systems.Large language models (LLMs) reflect societal norms and gender stereotypes in emotion attribution. This study investigates whether emotions are gendered and whether these variations are based on societal stereotypes. Five state-of-the-art LLMs (open- and closed-source) were tested to determine if they exhibit gendered emotions. The models were prompted to adopt a gendered persona and attribute emotions to an event, such as "When I had a serious argument with a dear person." The results show that all models consistently exhibit gendered emotions, influenced by gender stereotypes. Women are associated with SADNESS, while men are associated with ANGER. These findings align with established research in psychology and gender studies. The study highlights the complex societal interplay between language, gender, and emotion. The reproduction of emotion stereotypes in LLMs allows for detailed analysis of these stereotypes but raises questions about the predictive use of LLMs for emotion applications. The study provides a quantitative analysis of over 200,000 emotion attributions across 7,000 events and two personas, spanning over 400 unique emotions. The results show that gendered stereotypes are prevalent across the models, with men more likely to be associated with ANGER and women with SADNESS. The study also includes a qualitative analysis of model explanations, revealing that models often rely on societal stereotypes when attributing emotions. The findings suggest that LLMs reflect and amplify gender stereotypes in emotion attribution, which has implications for the fairness and inclusiveness of NLP systems. The study calls for interdisciplinary collaboration to address gendered emotions in NLP systems.