Internet of Things (IoT) networks are based on advanced augmented reality-based applications for better services. These advanced networks are always at threat due to different disasters in an urban congested environment. Delay tolerant networks is one of the well-known solutions with various features. However, due to heavy data traffic, it suffered from congestion and selfish behavior issues. The existing solutions endeavored to mitigate selfishness and enforce nodes to cooperate which leads to congestion. The information concerning other nodes in the network is needed and to be maintained, stored, and shared. In this article, the issue of the occurrence of congestion while mitigating the selfish behavior of nodes is taken into account, especially for augmented reality (AR) and 6 G-based IoT networks. This article proposes a credit-based congestion aware reward-based scheme for enforcing nodes to cooperate to refrain from selfishness. The load-sharing technique handles the congestion in 6G/AR-based IoT networks where the messages are shifted to low or non-congested nodes rather than dropping them. Simulation results affirm the efficacy of the proposed scheme in terms of data delivery ratio, network overhead, energy consumption, and latency and outperform compare to the state-of-the-art schemes.