The article introduces new jackknife IV estimators that can handle many weak instruments and error heteroskedasticity in a cluster sample setting. These estimators effectively account for cluster-specific effects and included exogenous regressors while maintaining the re-centering property of the jackknife method. The proposed procedures are the first to provide consistent estimators under many weak instrument scenarios. Additionally, the estimators are shown to be asymptotically normal, and t-statistics based on them are consistent under fixed alternatives. Monte Carlo simulations demonstrate that these t-statistics outperform alternative jackknife IV procedures in controlling size in finite samples.