aec-bench
aec-bench
AEC-Bench: A Multimodal Dataset for Architecture, Engineering, and Construction
Section
What it covers
Overview
What the dataset contains
Task taxonomy
Scopes, task families, instance counts
Accessing the dataset
manifest.jsonl, prefetching files from URLs
License
Apache 2.0
Citation
BibTeX
Overview
AEC-Bench is a multimodal dataset of real-world Architecture, Engineering, and Construction (AEC) documents — construction drawings, floor… See the full description on the dataset page: https://huggingface.co/datasets/nomic-ai/aec-bench.release-model-rollouts
AEC-Bench Release Model Rollouts
This dataset contains model rollouts from the first AEC-Bench release evaluation suite: model answers, trial metadata, execution traces, and task taxonomy for generated Architecture, Engineering, and Construction benchmark tasks.
AEC-Bench is a Python platform for creating, running, and evaluating AI agents on realistic AEC tasks. The benchmark is described in AEC-Bench: A Multimodal Benchmark for Agentic Systems in Architecture, Engineering, and… See the full description on the dataset page: https://huggingface.co/datasets/aec-bench/release-model-rollouts.AECBench
🏗️ AECBench
🇺🇸 English | 🇨🇳 中文说明
Project Introduction
AECBench is an open-source large language model Architecture, Engineering & Construction (AEC) domain evaluation benchmark jointly released by East China Architectural Design & Research Institute Co., Ltd. (ECADI) of China Construction Group and Tongji University. This dataset aims to systematically evaluate large language models' (LLMs) knowledge mastery, understanding, reasoning… See the full description on the dataset page: https://huggingface.co/datasets/jackluoluo/AECBench.