This report documents the results and observations on the Coordinated Research Project (CRP) of the International Atomic Energy Agency (IAEA) on "Neutronics Benchmark of CEFR Start-Up Tests." The China Experimental Fast Reactor (CEFR) is a 65MWt sodium-cooled fast reactor with highly enriched uranium oxide fuels. The reactor achieved the first criticality in 2010, and series of start-up tests were conducted to measure various reactor physics parameters. In 2018, IAEA has launched the CRP for validation and qualification of member states' computation capabilities in the field of fast reactor simulation utilizing the measured data in the CEFR start-up test. Twenty-nine international organizations from eighteen member countries, including Argonne National Laboratory, have participated in the CRP.The neutronics benchmark consists of six experimental measurements, criticality per fuel loading, control rod worth, sodium void reactivity, temperature reactivity, subassembly swap reactivity, and foil activation measurements. Participants also agreed on a numerical benchmark for the evaluation of integral reactivity coefficients related to the safety feature of the CEFR core. The Chinese Institute of Atomic Energy (CIAE) provided the benchmark specifications, drawings, measured data, and experimental uncertainties.The CRP has been conducted based on two phases. The first phase was a blind benchmark without measured data, and the second phase was an open benchmark with measured data. In the open benchmark, all participants could access the measured data and the results from other participants. Both phases have been completed, and participants are writing the IAEA technology document, which will be published in 2022.In this work, the CEFR start-up test parameters were simulated using the Argonne Reactor Computation (ARC) code suite, i.e., MC 2 -3/TWODANT, DIF3D-VARIANT, and PERSENT. The input preparation was performed using the PyARC script to generate input files for each case.The whole core was simulated using the region and material dependent 33-group cross sections. The results are generally well agreed with the measurements. For instance, the calculated core multiplication factor for the given CEFR critical configuration is 0.99890, which is 110 pm lower than the criticality. Other parameters, such as control rod worth, sodium void worth, temperature reactivity coefficient, subassembly swap reactivity, and foil activation rates, are aggreged with the measured values within the one standard deviation of the experimental uncertainties. A noticeable discrepancy between calculation and measurement was observed in the Au-197 (n,) reaction rates in the radial reflector and axial blankets. The primary reason for the discrepancy was poor modeling of the spatial self-shielding effect of the ) reaction in the epithermal energy range. In this work, the discrepancy was resolved by modeling the spatial self-shielding effect in the MC 2 -3 calculations and refining the cross-section group structures.The CRP was a good opportunity...