Task automation has been greatly empowered by the recent advances in Large Language Models (LLMs) via Python code, where the tasks range from software engineering development to general-purpose reasoning. While current benchmarks have shown that LLMs can solve tasks using programs like human developers, the majority of their evaluations are limited to short and self-contained algorithmic tasks or standalone function calls. Solving challenging and practical tasks requires the capability of utilizing **diverse function calls as tools** to efficiently implement functionalities like data analysis and web development. In addition, using multiple tools to solve a task needs compositional reasoning by accurately understanding **complex instructions**. Fulfilling both of these characteristics can pose a great challenge for LLMs. To assess how well LLMs can solve challenging and practical tasks via programs, we introduce BigCodeBench, a benchmark that challenges LLMs to invoke multiple function calls as tools from 139 libraries and 7 domains for 1,140 fine-grained tasks. To evaluate LLMs rigorously, each task encompasses 5.6 test cases with an average branch coverage of 99%. In addition, we propose a natural-language-oriented variant of BigCodeBench, BigCodeBench-Instruct, that automatically transforms the original docstrings into short instructions containing only essential information. Our extensive evaluation of 60 LLMs shows that **LLMs are not yet capable of following complex instructions to use function calls precisely, with scores up to 60%, significantly lower than the human performance of 97%**. The results underscore the need for further advancements in this area.

With increasing global passenger numbers, train stations face challenges in accommodating higher volumes while ensuring safety. Urbanization exacerbates this issue in densely populated, built-up areas with limited space. This poses increased safety risks due to potential overcrowding, especially during emergency evacuations. Therefore, integrating pedestrian simulators into early planning stages is crucial, yet time-consuming and manual, hindering thorough exploration of building design variants. In this article, we propose an interactive solution using a Detection Transformer to immediately assess the impact of design decisions on pedestrian flow and safety. Specifically, we present a digital train station model utilizing parametric Building Information Modeling (BIM) to generate many hundreds of floorplan variants to train our neural network. We then connect the trained network to a BIM authoring software, which allows for immediate evaluation of design changes. To address the diverse design parameter space, we present distinct customization options for the network to integrate supplementary inputs, merging simulator and floorplan data for accurate predictions. Notably, our approach conforms to official train station design guidelines and congestion standards, demonstrating its applicability in practice and ensuring compliance with safety regulations. To showcase the effectiveness and versatility of our methodology, we apply it to a real-world use case.