Human plasma mirrors an individual´s health state and is routinely collected during clinical care for diagnostic purposes. As a potentially rich source of biomarkers, plasma continues to be a primary focus of biomedical and clinical pursuits until today. Due to its unbiased nature, liquid chromatography-mass spectrometry (LC-MS)-based proteomics has become a key method for the discovery of novel plasma biomarkers. However, MS-based approaches are challenged by the high dynamic range of plasma proteins. In the present study, we generated a multispecies sample set based on a human tryptic plasma digest which contains varying spike-in levels of tryptically digested yeast and E. coli (PYE) proteomes addressing the challenges of a high protein dynamic range. With the aim to benchmark and evaluate the quantitative performance of neat plasma analysis, the sample set was distributed and analysed on various state-of-the-art LC-MS platforms across twelve different sites. The resulting data set encompasses in total 1,116 individual LC-MS runs and was centrally analyzed. Data sets derived from different setups were compared in terms of identifications, data completeness, accuracy, precision, as well as reproducibility. The acquisition mode was one of the main factors impacting proteome coverage, reproducibility and data completeness. Beside the acquisition mode, other parameters, such as gradient length, cycle time, instrument settings etc. affected proteomic depth as well as accuracy and precision of the quantitative measurements. The present dataset poses a valuable resource for further optimizing LC-MS workflows with the aim to enhance the accuracy and reproducibility of plasma proteome analysis, crucial for clinical biomarker discovery and verification.