The Best Automated Social Science: Language Models as Scientist and Subjects NBER Alternatives in 2026

SpeechMap.AI compares language models’ refusal rates on controversial speech topics.

OpenLM Arena compares open language models through leaderboards and benchmark results.

VoxelBench benchmarks language models on creating voxel builds from prompts.

KNIME provides data analysis tools and a visual workflow builder for data science solutions.

The Hugging Face Open LLM Leaderboard compares language models across evaluation tests including IFEval, MATH, and MMLU-Pro.

FastDatasets helps create training datasets for large language models.

Automatic "Differentiation" via Text, using large language models to backpropagate textual gradients.

LLMeBench provides a flexible framework for benchmarking large language models.

BambooAI analyzes data with a language model in a persistent Python kernel and supports web and Docker workflows.

Uncover LLM evaluation's importance and explore methods for assessing its performance and impact across industries.

Powerdrill Bloom presents an AI workspace with memory for data analysis and agents.

An open-source visual programming environment for battle-testing prompts to LLMs.

oneDAL is a C++ and DPC++ library for accelerated machine learning routines on tabular data, governed by the UXL Foundation.

Materials and IPython notebooks for "Python for Data Analysis" by Wes McKinney, published by O'Reilly Media.

This project profiles data quality and supports exploratory analysis of Pandas and Spark DataFrames.

Optimus supports agile data preparation workflows across Pandas, Dask, cuDF, Vaex, and PySpark.

An AI-first alternative to Jupyter, Deepnote supports Python, R, and SQL, with local IDE use, cloud collaboration, and data apps.

ADAM analyzes genomics data with specialized file formats built on Apache Spark, Avro, and Parquet.

Lance provides an open lakehouse format for multimodal AI, with vector indexing and data versioning.

Cleanlab provides a Python library for data quality and machine learning with messy real world data and labels.

Desbordante profiles data to find patterns and supports data-cleaning scenarios through console and web application interfaces.

AgentBench evaluates large language models acting as agents through a research paper and discussion page.

DrivenData hosts data science competitions where participants build predictive models for organizations addressing social challenges.

Indico benchmarks language models on extractive, classification, and predictive tasks.