Worked on enhancing the Vsibench metric calculation within the EvolvingLMMs-Lab/lmms-eval repository, focusing on expanding its flexibility for benchmarking tasks. Introduced support for functools.partial in Python, allowing users to configure metric calculations with partial functions while preserving compatibility with existing workflows. This update enables more customizable evaluation pipelines and facilitates faster experimentation for teams integrating lmms-eval into their processes. The work centered on Python programming and data analysis, ensuring that new features could be adopted smoothly without disrupting established usage patterns. No bug fixes were recorded during this period, with efforts concentrated on delivering this targeted feature enhancement.