The paper aims to assess how Polish enterprises work together on innovative projects as part of business clusters and otherwise, and it also evaluates the impact that this cooperation has on the overall performance of Polish enterprises in terms of innovation. Additionally, the authors seek to identify the impact of European Union funds intended for innovative projects on the cooperation of Polish enterprises in innovation. The data comes from the 2008–2010 Community Innovation Survey conducted by Poland’s Central Statistical Office (GUS) in 2011. The sample covers 7,783 medium-sized and large manufacturing firms, with n=80 big cluster-based firms singled out for the final study. The results of Structural Equation Modeling (SEM), with an additional analysis of the distribution of the estimation errors with multiple sampling with replacement from the sample (non-parametric bootstrap method), show a statistically significant relationship between “non-cluster innovation cooperation with international partners” (suppliers, customers, competitors, capital group members and institutional partners) and innovation performance measured by the log of fraction of sales of innovative products in total sales, the authors say. They add that a statistically significant relationship also exists in the case of “cluster innovation cooperation with domestic partners.” Moreover, the authors argue that the relationship between cluster innovation cooperation, with both domestic and international partners, and non-cluster innovation cooperation demonstrates a “cooperation learning effect.” The impact of European Union funds is visible only in the case of cluster cooperation, testifying to “the existence of cluster cooperation additionality,” the authors conclude.