This paper introduces the Context-based NeRF Compression (CNC) framework, which aims to reduce the storage size of Neural Radiance Field (NeRF) models while maintaining high fidelity and rendering speed. The authors leverage context models to minimize the entropy of explicit feature embeddings, which are a significant portion of NeRF's storage. Specifically, they design level-wise and dimension-wise context models to capture both multi-level and cross-dimensional dependencies in the feature embeddings. Additionally, they exploit hash collision and occupancy grids to enhance the accuracy of context modeling. The proposed CNC framework achieves a 100× and 70× reduction in storage size compared to the baseline Instant-NGP on Synthetic-NeRF and Tanks and Temples datasets, respectively, while improving fidelity. The method outperforms the state-of-the-art NeRF compression algorithm, BiRF, with an 86.7% and 82.3% reduction in storage size. The code for the proposed method is available at <https://github.com/YihangChen-ee/CNC>.This paper introduces the Context-based NeRF Compression (CNC) framework, which aims to reduce the storage size of Neural Radiance Field (NeRF) models while maintaining high fidelity and rendering speed. The authors leverage context models to minimize the entropy of explicit feature embeddings, which are a significant portion of NeRF's storage. Specifically, they design level-wise and dimension-wise context models to capture both multi-level and cross-dimensional dependencies in the feature embeddings. Additionally, they exploit hash collision and occupancy grids to enhance the accuracy of context modeling. The proposed CNC framework achieves a 100× and 70× reduction in storage size compared to the baseline Instant-NGP on Synthetic-NeRF and Tanks and Temples datasets, respectively, while improving fidelity. The method outperforms the state-of-the-art NeRF compression algorithm, BiRF, with an 86.7% and 82.3% reduction in storage size. The code for the proposed method is available at <https://github.com/YihangChen-ee/CNC>.