This dataset consists of ten different parts designed for manufacturing on a three-axis CNC milling machine with gantry design (or milling machines with more degrees of freedom). All parts can be made from a raw material block of 80 × 80 × 50 mm using a single end mill (maximum diameter 6 mm) without reclamping. The parts' geometries are designed to include an increasing number of features and rising complexity. The dataset contains, among others, slots, holes, rectangular and circular outer and inner contours, pockets, and freeform surfaces. These parts do not represent any functional parts, but purely geometries intended for the generation of manufacturing programs and the analysis of machining processes. The dataset extends the benchmark part introduced by Mertes et al. (2024) for the evaluation of form and positional tolerances (https://doi.org/10.1016/j.dte.2024.100024), which is based on ISO 10791-7. This part is also included in this collection (Part No. 8). The following table provides an overview of the key features of each part. Part No.: Key features: 1 Slots 2 4 holes 3 4 holes, surrounding step (rectangular) 4 4 holes, surrounding step (circular) 5 4 holes, surrounding step (semi-circular/semi-rectangular) 6 4 holes, surrounding step (rectangular), circular pocket 7 Square pocket with negative freeform 8 Multiple inclined surrounding steps (circular and rectangular) 9 4 holes, circular pocket, multiple negative freeforms 10 4 holes, multiple inclined surrounding steps (circular and rectangular), multiple pockets (circular and rectangular), negative freeform The models were used for G-code generation and for analyzing the resulting motion profiles. They were also manufactured on a three-axis CNC machine tool. The repository contains the corresponding STEP files, generated G-codes, and a collection of rendered images of the manufactured parts. This dataset thus provides a versatile basis for research, education, and applications in manufacturing technology, CNC programming, and process analysis, as well as for generating training data in machine learning. License: This dataset is distributed under the Creative Commons Attribution (CC-BY) license. The data may be freely used, modified, and shared, provided that the original authors are properly credited. Contact:Dipl.-Ing. Marius SchmitzRPTU Kaiserslautern-Landaumarius.schmitz@rptu.de