datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
or-bench
OR-Bench: An Over-Refusal Benchmark for Large Language Models
Please see our demo at HuggingFace Spaces.
Overall Plots of Model Performances
Below is the overall model performance. X axis shows the rejection rate on OR-Bench-Hard-1K and Y axis shows the rejection rate on OR-Bench-Toxic. The best aligned model should be on the top left corner of the plot where the model rejects the most number of toxic prompts and least number of safe prompts. We also plot a blue line… See the full description on the dataset page: https://huggingface.co/datasets/bench-llm/or-bench.wildjailbreak
WildJailbreak Dataset Card
WildJailbreak is an open-source synthetic safety-training dataset with 262K vanilla (direct harmful requests) and adversarial (complex adversarial jailbreaks) prompt-response pairs. In order to mitigate exaggerated safety behaviors, WildJailbreaks provides two contrastive types of queries: 1) harmful queries (both vanilla and adversarial) and 2) benign queries that resemble harmful queries in form but contain no harmful intent.
Vanilla Harmful: direct… See the full description on the dataset page: https://huggingface.co/datasets/allenai/wildjailbreak.JailBreakV-28k
⛓💥 JailBreakV-28K: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks
🌐 GitHub | 🛎 Project Page ｜ 👉 Download full datasets
If you like our project, please give us a star ⭐ on Hugging Face for the latest update.
📰 News
Date
Event
2024/07/09
🎉 Our paper is accepted by COLM 2024.
2024/06/22
🛠️ We have updated our version to V0.2, which supports users to customize their attack models… See the full description on the dataset page: https://huggingface.co/datasets/JailbreakV-28K/JailBreakV-28k.rtx-5090-benchmarks
RTX 5090 LLM Benchmarks
Speed and quality benchmarks for quantized LLMs on NVIDIA RTX 5090 32GB, measured with llm-bench-rig.
Quality Benchmarks
Generative evaluation through llama-server chat completions. Replicates standard benchmark methodology using custom evaluators — no lm-evaluation-harness dependency.
Results are split by reasoning mode: comparing a thinking-on (reasoning) model's quality against a thinking-off model is apples-to-oranges, so the two groups… See the full description on the dataset page: https://huggingface.co/datasets/witcheer/rtx-5090-benchmarks.paradetox
ParaDetox: Text Detoxification with Parallel Data (English)
This repository contains information about ParaDetox dataset -- the first parallel corpus for the detoxification task -- as well as models and evaluation methodology for the detoxification of English texts. The original paper "ParaDetox: Detoxification with Parallel Data" was presented at ACL 2022 main conference.
📰 Updates
[2025] !!!NOW OPEN!!! TextDetox CLEF2025 shared task: for even more -- 15 languages! website… See the full description on the dataset page: https://huggingface.co/datasets/s-nlp/paradetox.or-bench
OR-Bench: An Over-Refusal Benchmark for Large Language Models
Please see our demo at HuggingFace Spaces.
Overall Plots of Model Performances
Below is the overall model performance. X axis shows the rejection rate on OR-Bench-Hard-1K and Y axis shows the rejection rate on OR-Bench-Toxic. The best aligned model should be on the top left corner of the plot where the model rejects the most number of toxic prompts and least number of safe prompts. We also plot a blue line… See the full description on the dataset page: https://huggingface.co/datasets/bench-llms/or-bench.Tunisian-Proverbs-with-Image-Associations-A-Cultural-and-Linguistic-DatasetTunisian Proverbs with Image Associations: A Cultural and Linguistic Dataset
Description
This dataset explores the rich oral tradition of Tunisian proverbs mapped into text format, pairing each with contextual explanations, English translations both word-to-word and it's equivalent Target Language dynamic, Automated prompt and AI-generated visual interpretations.
It bridges linguistic, cultural, and visual modalities making it valuable for tasks in cross-cultural NLP, generative… See the full description on the dataset page: https://huggingface.co/datasets/HabibaAbderrahim/Tunisian-Proverbs-with-Image-Associations-A-Cultural-and-Linguistic-Dataset.BiasShadesInterested in contributing? Speak a language not represented here? Disagree with an annotation? Please submit feedback in the Community tab!
Dataset Card for BiasShades
Note: This dataset may NOT be used as training data in any form (pre-training, fine-tuning, post-training, etc.) without express permission from creators.
Dataset Details
Version: 1.0
License: SHADES 1 Montreal Data License
Dataset Description
728 stereotypes and associated… See the full description on the dataset page: https://huggingface.co/datasets/LanguageShades/BiasShades.crypto-news-coindesk-2020-2025
CoinDesk Cryptocurrency News Dataset (2020–2025)
This dataset contains cryptocurrency-related news articles sourced from CoinDesk Data, accessed programmatically via the CryptoCompare API. The dataset is curated and published for academic and research purposes, with a focus on analyzing the relationship between news and cryptocurrency market dynamics.
Time Period
January 1, 2020 – January 1, 2025
Content Overview
Each record in the dataset… See the full description on the dataset page: https://huggingface.co/datasets/maryamfakhari/crypto-news-coindesk-2020-2025.or-bench
OR-Bench: An Over-Refusal Benchmark for Large Language Models
Please see our leaderboard at HuggingFace Spaces.
Overall Plots of Model Performances
Below is the overall model performance. X axis shows the rejection rate on OR-Bench-Hard-1K and Y axis shows the rejection rate on OR-Bench-Toxic. The best aligned model should be on the top left corner of the plot where the model rejects the most number of toxic prompts and least number of safe prompts. We also plot a blue… See the full description on the dataset page: https://huggingface.co/datasets/orbench-llm/or-bench.morphogen
MORPHOGEN: A Multilingual Benchmark for Evaluating Gender-Aware Morphological Generation
This repository contains the MORPHOGEN dataset introduced in our ACL 2026 paper: "MORPHOGEN: A Multilingual Benchmark for Evaluating Gender-Aware Morphological Generation".
Introduction
MORPHOGEN is a morphologically grounded, large-scale benchmark designed to evaluate the gender-aware generation capabilities of Large Language Models (LLMs) in three typologically diverse languages:… See the full description on the dataset page: https://huggingface.co/datasets/ag2003/morphogen.social-instagram-marketing
Social — Instagram Marketing Multimodal Dataset
A synthetic, multimodal dataset for Social, an AI Instagram-marketing agent. Every row is a single Instagram post idea that pairs a marketing caption with a matching AI-generated image, conditioned on a business brief and brand preferences.
Agent pattern: owner brief + brand preferences → 3 similar successful posts (retrieval / recommendation) + 1 freshly generated post (caption + image).
Rows (total)
1,447… See the full description on the dataset page: https://huggingface.co/datasets/avihayamor/social-instagram-marketing.or-bench-toxic-all
OR-Bench: An Over-Refusal Benchmark for Large Language Models
This dataset constains highly toxic prompts, use with caution!!!
Please see our demo at HuggingFace Spaces.
Overall Plots of Model Performances
Below is the overall model performance. X axis shows the rejection rate on OR-Bench-Hard-1K and Y axis shows the rejection rate on OR-Bench-Toxic. The best aligned model should be on the top left corner of the plot where the model rejects the most number of toxic… See the full description on the dataset page: https://huggingface.co/datasets/bench-llms/or-bench-toxic-all.yc-companies-august-2025
Y Combinator Companies Dataset
Dataset Description
This dataset contains information about 5,404 Y Combinator funded companies that have been publicly launched, sourced from the YC-OSS-API.
Dataset Summary
Total Companies: 5,404
Time Range: Summer 2005 - Summer 2025
Update Frequency: Snapshot from August 2025
Source: YC-OSS-API
Dataset Structure
Data Fields
id: Unique identifier for each company
name: Company name… See the full description on the dataset page: https://huggingface.co/datasets/jeffboudier/yc-companies-august-2025.ChEBI-20-MM
ChEBI-20-MM Dataset
Overview
The ChEBI-20-MM is an extensive and multi-modal benchmark developed from the ChEBI-20 dataset. It is designed to provide a comprehensive benchmark for evaluating various models' capabilities in the field of molecular science. This benchmark integrates multi-modal data, including InChI, IUPAC, SELFIES, and images, making it a versatile tool for a wide range of molecular tasks.
Dataset Description
ChEBI-20-MM is an expansion of the… See the full description on the dataset page: https://huggingface.co/datasets/liupf/ChEBI-20-MM.rtx-5090-benchmarks
RTX 5090 LLM Benchmarks
Speed and quality benchmarks for quantized LLMs on NVIDIA RTX 5090 32GB, measured with llm-bench-rig.
Quality Benchmarks
Generative evaluation through llama-server chat completions. Replicates standard benchmark methodology using custom evaluators — no lm-evaluation-harness dependency.
Results are split by reasoning mode: comparing a thinking-on (reasoning) model's quality against a thinking-off model is apples-to-oranges, so the two groups… See the full description on the dataset page: https://huggingface.co/datasets/omegaprime669/rtx-5090-benchmarks.property-pilot-tickets
🏢 PropertyPilot — Maintenance Tickets
A synthetic dataset of 13,725 residential-maintenance tickets written the way real tenants write them — polite, panicked, passive-aggressive, or confused — each paired with operational metadata (category, urgency, assigned contractor, cost, resolution time).
Built for an end-to-end NLP pipeline: triage classification, similar-case retrieval (embeddings + FAISS), and work-order / reply generation.
About this release. Earlier versions of… See the full description on the dataset page: https://huggingface.co/datasets/propertypilot/property-pilot-tickets.HumanCreativityBenchmark
The Human Creativity Benchmark (HCB)
Expert evaluations of AI-generated creative work, built to separate two signals that single-score benchmarks collapse: convergence, where professionals align around shared, checkable standards, and divergence, where creative taste legitimately differs. Each AI output is judged by domain professionals through three complementary lenses — forced-choice pairwise comparisons, 1-5 scalar ratings on prompt adherence, usability, and visual appeal… See the full description on the dataset page: https://huggingface.co/datasets/contralabs/HumanCreativityBenchmark.phenotype-catalog
Ethnic Erotic Phenotype Catalog
A structured complement to Wikipedia for ethnographic data — 1,700+ ethnic groups indexed with normalized linguistic, geographic, cultural, and phenotype metadata, plus 23K+ notable-people references and 5K+ vision-grounded per-image phenotype observations.
Curated from the live catalog at ethnicerotic.com and published as an open dataset for anthropological reference, AI training, and ethnographic research.
What's in v6
Two columns… See the full description on the dataset page: https://huggingface.co/datasets/EthnicErotic/phenotype-catalog.ru_paradetox
ParaDetox: Text Detoxification with Parallel Data (Russian)
This repository contains information about Russian Paradetox dataset -- the first parallel corpus for the detoxification task -- as well as models for the detoxification of Russian texts.
📰 Updates
[2025] !!!NOW OPEN!!! TextDetox CLEF2025 shared task: for even more -- 15 languages! website 🤗Starter Kit
[2025] COLNG2025: Daryna Dementieva, Nikolay Babakov, Amit Ronen, Abinew Ali Ayele, Naquee Rizwan, Florian Schneider… See the full description on the dataset page: https://huggingface.co/datasets/s-nlp/ru_paradetox.Myanmar-Tuberculosis-Guidelines-Instructions
Myanmar Tuberculosis Guidelines Instructions
A bilingual instructional dataset built to support Myanmar's ongoing fight against tuberculosis — turning life-saving guidelines into a usable resource for healthcare workers, educators, and AI researchers working with low-resource languages.
Authors: Min Si Thu, Khin Myat Noe
Abstract
Tuberculosis is still one of Myanmar's biggest public health problems. Part of the difficulty is that good, standardized TB education… See the full description on the dataset page: https://huggingface.co/datasets/jojo-ai-mst/Myanmar-Tuberculosis-Guidelines-Instructions.JailBreakV-28k
⛓💥 JailBreakV-28K: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks
🌐 GitHub | 🛎 Project Page ｜ 👉 Download full datasets
If you like our project, please give us a star ⭐ on Hugging Face for the latest update.
📰 News
Date
Event
2024/07/09
🎉 Our paper is accepted by COLM 2024.
2024/06/22
🛠️ We have updated our version to V0.2, which supports users to customize their attack models… See the full description on the dataset page: https://huggingface.co/datasets/Ngixdev/JailBreakV-28k.UIS-QA
UIS-QA: A Benchmark for Unindexed Information Seeking
Figure 1. UIS problem. Standard agents (bottom) rely on indexed information and often fail or hallucinate; UIS-capable agents (top) use additional tools to excavate unindexed information and solve UIS tasks.
If .figs do not load, see the paper.
🔔 News
[2026.03.10] 🎉 We release the UIS-QA dataset and the paper (ICLR 2026, arXiv) today!
📋 Dataset Description
Homepage
Paper… See the full description on the dataset page: https://huggingface.co/datasets/UIS-Digger/UIS-QA.vwp
Dataset Card for Visual Writing Prompts Dataset (VWP)
Website | Github Repository | arXiv e-Print
The Visual Writing Prompts (VWP) dataset contains almost 2K selected sequences of
movie shots, each including 5-10 images. The image sequences are aligned with a total of 12K stories which are collected via crowdsourcing given the image sequences and up to 5 grounded characters from the corresponding image sequence.
Dataset Details
Dataset Links
TACL… See the full description on the dataset page: https://huggingface.co/datasets/tonyhong/vwp.emotion-prediction-comet-atomic-2020
emotion-prediction-comet-atomic-2020
This dataset extends the COMET-Atomic-2020 commonsense reasoning dataset by focusing on the xReact (subject’s emotional reaction) and oReact (other person’s emotional reaction) relations.
Description
Each entry is expanded into a realistic, three‑sentence scenario, replacing the placeholders (PersonX/PersonY) with human names and adding contextual details.
Source (comet-atomic-2020) Example:
source
relation
target
PersonX… See the full description on the dataset page: https://huggingface.co/datasets/id4thomas/emotion-prediction-comet-atomic-2020.or-bench
OR-Bench: An Over-Refusal Benchmark for Large Language Models
Please see our demo at HuggingFace Spaces.
Overall Plots of Model Performances
Below is the overall model performance. X axis shows the rejection rate on OR-Bench-Hard-1K and Y axis shows the rejection rate on OR-Bench-Toxic. The best aligned model should be on the top left corner of the plot where the model rejects the most number of toxic prompts and least number of safe prompts. We also plot a blue… See the full description on the dataset page: https://huggingface.co/datasets/jerogo/or-bench.DALL-E-Prompts-OpenAI-ChatGPT
Dataset Card for Dataset Name
Dataset Summary
This dataset has been generated using Prompt Generator for OpenAI's DALL-E.
Languages
English
Dataset Structure
1.000.000 Prompts
MCiteBench
MCiteBench Dataset
MCiteBench is a benchmark for evaluating the ability of Multimodal Large Language Models (MLLMs) to generate text with citations in multimodal contexts.
Websites: https://caiyuhu.github.io/MCiteBench
Paper: https://arxiv.org/abs/2503.02589
Code: https://github.com/caiyuhu/MCiteBench
Data Download
Please download the MCiteBench_full_dataset.zip. It contains the data.jsonl file and the visual_resources folder.
Data Statistics… See the full description on the dataset page: https://huggingface.co/datasets/caiyuhu/MCiteBench.dou-brazil-dataset
Dataset Card for Dataset Diário Oficial da União (DOU)
The Diário Oficial da União (DOU) is the official government gazette of Brazil, published by the National Press. It serves as the primary means of communication for federal government acts, including laws, decrees, ordinances, public notices, and other official decisions. The DOU ensures transparency and legal validity for government actions and is divided into three sections:
Section 1: Publishes laws, decrees, and… See the full description on the dataset page: https://huggingface.co/datasets/gerson-vfs/dou-brazil-dataset.IKEA-Home-Decor-Furniture-Dataset
IKEA Home Decor & Furniture Product Dataset
A rich, structured dataset of IKEA home decor and furniture products, featuring deep category taxonomies, full product descriptions, measurements, features, and image URLs. Ideal for training product recommendation models, interior design AI applications, multimodal models, and e-commerce search systems.
Dataset Overview
Field
Details
Source
IKEA (multi-country)
Total Records
400+
Category Focus
Home… See the full description on the dataset page: https://huggingface.co/datasets/crawlfeeds/IKEA-Home-Decor-Furniture-Dataset.