Secondary ion mass spectrometry (SIMS) is often used to determine the sulfur contents and isotope ratios of metallic alloys in meteorites or high‐pressure experimental samples. However, SIMS analyses involve calibration and the determination of instrumental mass fractionation in reference materials with a matrix composition similar to that of the unknown samples. To provide metallic reference materials adapted to S measurements via SIMS, we synthesised a series of twenty‐eight alloys comprising four FeNi(±Si) compositions (Fe95Ni5, Fe90Ni10, Fe80Ni20, and Fe80Ni15Si5) with S contents varying from 100 μg g−1 to 4 g/100g using the “melt spinning” method, which guarantees that the metal alloys are rapidly quenched at ~ 106 K s−1. Sulfur contents were determined at the Service d'Analyse des Roches et Minéraux at the CRPG and absolute δ34S values were determined by multi‐collector ICP‐MS (MC‐ICP‐MS, ThermoScientific Neptune) and isotope ratio mass spectrometry (Thermoscientific Delta V). A δ34S value of 16.01 ± 0.31‰ was consistently obtained using the MC‐ICP‐MS, which was indistinguishable of the δ34S value of the FeS starting material (15.95 ± 0.08‰). It suggests that S did not undergo isotopic fractionation during the melting process. Of fifteen samples containing ≤ 5000 μg g−1 S, SIMS measurements with 15‐μm‐diameter spots were repeatable to within 10% relative (1 standard deviation, 1s) for S contents and 2‰ for δ34S values. However, samples containing > 5000 μg g−1 S showed FeNi–FeS immiscibility, leading to minor dispersion of the S mass fractions and δ34S values. No matrix effect was observed for Fe‐Ni, Si, or S contents in terms of the calibration curves and instrumental mass fractionation. We ultimately recommend eight samples as reliable reference materials for S isotopic measurements by SIMS, which we can share worldwide with other laboratories.