McLean, ReginaldReginaldMcLeanChatzaroulas, EvangelosEvangelosChatzaroulasMcCutcheon, LucLucMcCutcheonRöder, FrankFrankRöderHe, ZhanpengZhanpengHeYu, TianheTianheYuJulian, RyanRyanJulianZentner, K. R.K. R.ZentnerTerry, JordanJordanTerryWoungang, IsaacIsaacWoungangFarsad, NarimanNarimanFarsadCastro, Pablo SamuelPablo SamuelCastro2026-02-172026-02-172025-11-2139th Conference on Neural Information Processing Systems, NeurIPS 2025https://hdl.handle.net/11420/61570Meta-World is widely used for evaluating multi-task and meta-reinforcement learning agents, which are challenged to master diverse skills simultaneously. Since its introduction however, there have been numerous undocumented changes which inhibit a fair comparison of algorithms. This work strives to disambiguate these results from the literature, while also leveraging the past versions of Meta-World to provide insights into multi-task and meta-reinforcement learning benchmark design. Through this process we release a new open-source version of Meta-World1 that has full reproducibility of past results, is more technically ergonomic, and gives users more control over the tasks that are included in a task set.enhttps://creativecommons.org/licenses/by/4.0/Computer Science, Information and General Works::005: Computer Programming, Programs, Data and SecurityComputer Science, Information and General Works::006: Special computer methodsPreprinthttps://doi.org/10.15480/882.1672310.48550/arXiv.2505.1128910.15480/882.167232505.11289Preprint