datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
soc-ratchakitcha
Royal Gazette Thailand (Ratchakitcha) Dataset
ชุดข้อมูลราชกิจจานุเบกษา (แบบ Machine Readable)
โครงการ Open Law Data Thailand ร่วมกับคณะกรรมาธิการการพาณิชย์และการอุตสาหกรรม วุฒิสภา ได้รับความอนุเคราะห์ข้อมูลจาก สำนักเลขาธิการคณะรัฐมนตรี (สลค.) เพื่อเผยแพร่ข้อมูลกฎหมายไทยสู่สาธารณะในรูปแบบที่ประมวลผลได้ด้วยคอมพิวเตอร์ (Machine Readable) เพื่อส่งเสริมนวัตกรรม Legal Tech และ AI ของประเทศไทย
Dataset Description
ชุดข้อมูลนี้รวบรวมรายการประกาศในราชกิจจานุเบกษา… See the full description on the dataset page: https://huggingface.co/datasets/open-law-data-thailand/soc-ratchakitcha.medical-o1-reasoning-SFT
News
[2025/04/22] We split the data and kept only the medical SFT dataset (medical_o1_sft.json). The file medical_o1_sft_mix.json contains a mix of medical and general instruction data.
[2025/02/22] We released the distilled dataset from Deepseek-R1 based on medical verifiable problems. You can use it to initialize your models with the reasoning chain from Deepseek-R1.
[2024/12/25] We open-sourced the medical reasoning dataset for SFT, built on medical verifiable problems and an… See the full description on the dataset page: https://huggingface.co/datasets/FreedomIntelligence/medical-o1-reasoning-SFT.FlashRAG_datasets
⚡FlashRAG: A Python Toolkit for Efficient RAG Research
FlashRAG is a Python toolkit for the reproduction and development of Retrieval Augmented Generation (RAG) research. Our toolkit includes 36 pre-processed benchmark RAG datasets and 16 state-of-the-art RAG algorithms.
With FlashRAG and provided resources, you can effortlessly reproduce existing SOTA works in the RAG domain or implement your custom RAG processes and components.
For more information, please view our GitHub repo… See the full description on the dataset page: https://huggingface.co/datasets/RUC-NLPIR/FlashRAG_datasets.RWKU
Dataset Card for Real-World Knowledge Unlearning Benchmark (RWKU)
Dataset Summary
RWKU is a real-world knowledge unlearning benchmark specifically designed for large language models (LLMs).
This benchmark contains 200 real-world unlearning targets and 13,131 multi-level forget probes, including 3,268 fill-in-the-blank probes, 2,879 question-answer probes, and 6,984 adversarial-attack probes.
RWKU is designed based on the following three key factors:
For the task setting… See the full description on the dataset page: https://huggingface.co/datasets/jinzhuoran/RWKU.t2-ragbench
Dataset Card for T2-RAGBench
Project Page | Paper | Code
IMPORTANT NOTICE:
We deleted VQAonBD from the dataset due to low quality of the question reformulations. If you still want to use it you will find the data in the previous commit history.
Dataset Description
Dataset Summary
T2-RAGBench is a benchmark dataset designed to evaluate Retrieval-Augmented Generation (RAG) on financial documents containing both text and tables. It consists of 23,088… See the full description on the dataset page: https://huggingface.co/datasets/G4KMU/t2-ragbench.newswire
Dataset Card for NewsWire
Dataset Summary
NewsWire contains 2.7 million unique public domain U.S. news wire articles, written between 1878 and 1977. Locations in these articles are georeferenced, topics are tagged using customized neural topic classification, named entities are recognized, and individuals are disambiguated to Wikipedia using a novel entity disambiguation model.
Languages
English (en)
Dataset Structure
Each year in the dataset is… See the full description on the dataset page: https://huggingface.co/datasets/dell-research-harvard/newswire.Fable-5.1-Max-Reasoning-Filtered-10000x
Dataset Description
This dataset contains 10,000 agentic coding and reasoning multi-turn high-quality traces generated by the new Fable 5.1 model using max reasoning effort.
It holds almost 500,000,000 tokens of step-by-step chain-of-thought programming across multiple complex domains.
It has also been deduplicated and heavily filtered to remove low-quality traces, keeping only high-quality traces.
Dataset Statistics
Metric
Value
Total Examples
10,000… See the full description on the dataset page: https://huggingface.co/datasets/MoreThought/Fable-5.1-Max-Reasoning-Filtered-10000x.R1-Code-Interpreter-Data
R1-Code-Interpreter: Training LLMs to Reason with Code via Supervised and Reinforcement Learning
Our code is based on Llama-factory/VeRL/Search-R1 for the SFT and RL training and SymBench/BIG-Bench-Hard/reasoning-gym for datasets/benchmarks of reasoning/planning tasks.
📝 Introduction
R1-Code-Interpreter is the first framework to train LLMs for step-by-step code reasoning using multi-turn supervised fine-tuning and reinforcement learning. By curating 144 diverse… See the full description on the dataset page: https://huggingface.co/datasets/yongchao98/R1-Code-Interpreter-Data.riddleIt contains 585 English riddles. The top 173 was adjusted by GPT4.
Real-3DQA
Real-3DQA
Do 3D Large Language Models Really Understand 3D Spatial Relationships?
🌐 Project Page · 📄 Paper · 💻 GitHub
Overview
Real-3DQA is a debiased 3D spatial QA benchmark with viewpoint rotation consistency evaluation. It addresses two key shortcomings of existing benchmarks:
Language Shortcut Filtering — Questions answerable through linguistic priors alone are removed by comparing 3D-LLMs against blind text-only counterparts.
Viewpoint Rotation Score (VRS) — Each… See the full description on the dataset page: https://huggingface.co/datasets/Oliver-Ma/Real-3DQA.riddles-qa
Riddles QA
This dataset contains 500 riddles and their answers.
Credits
The riddles were downloaded from https://github.com/nkilm/riddles-api
vsr_random
VSR: Visual Spatial Reasoning
This is the random set of VSR: Visual Spatial Reasoning (TACL 2023) [paper].
Usage
from datasets import load_dataset
data_files = {"train": "train.jsonl", "dev": "dev.jsonl", "test": "test.jsonl"}
dataset = load_dataset("cambridgeltl/vsr_random", data_files=data_files)
Note that the image files still need to be downloaded separately. See data/ for details.
Go to our github repo for more introductions.
Citation
If you find VSR… See the full description on the dataset page: https://huggingface.co/datasets/cambridgeltl/vsr_random.race-cRace-C : additional data for race (high school/middle school) but for college level
https://github.com/mrcdata/race-c
@InProceedings{pmlr-v101-liang19a,
title={A New Multi-choice Reading Comprehension Dataset for Curriculum Learning},
author={Liang, Yichan and Li, Jianheng and Yin, Jian},
booktitle={Proceedings of The Eleventh Asian Conference on Machine Learning},
pages={742--757},
year={2019}
}
ReClor
ReClor: A Reading Comprehension Dataset Requiring Logical Reasoning
This repository provides the dataset from the paper ReClor: A Reading Comprehension Dataset Requiring Logical Reasoning.
We corrected the original format issues to ensure full compatibility with the Hugging Face Datasets library.
For more details, please visit the original project page.
RHELM
RHELM: Beyond Static Dialogues
Benchmarking Realistic, Heterogeneous, and Evolving Long-Horizon Memory
RHELM is a benchmark for evaluating long-horizon memory capabilities in AI assistants.
Unlike benchmarks built around static dialogues, RHELM provides realistic,
heterogeneous, and temporally evolving memory sources, together with
challenging questions that require multi-hop reasoning, temporal synthesis, and
hallucination detection.
⚠️ All characters, events, and personal… See the full description on the dataset page: https://huggingface.co/datasets/microsoft/RHELM.refute
Can AI read new science honestly?
Models can sound convincing while misreading a result or expressing more confidence than the evidence deserves. That matters when people use them to summarize papers, compare studies, or decide what to investigate next.
REFUTE tests whether a model knows the finding, spots quiet flaws, names what would overturn a claim, and matches its confidence to the evidence.
Truth Score is the main result. It combines factual accuracy, flaw… See the full description on the dataset page: https://huggingface.co/datasets/BGPT-OFFICIAL/refute.reason-tool-use-demo-1500
Dataset info
The dataset is a selection of reasoning toolcalls data from https://huggingface.co/datasets/interstellarninja/hermes_reasoning_tool_use, which contains data from Hermes-Tools、Glaive-FC、ToolAce、Nvidia-When2Call.
The format has been transformed to adapt llama-factory v1 training pipeline.
Math-RL-Tasks
Ulam AI Math RL Tasks
Forty original, verifier-backed mathematical reasoning tasks packaged as ten
independent RL environments. The collection spans advanced graduate exercises,
research-style exact computation and structural generalization problems in
algebraic geometry, arithmetic geometry, combinatorics, topology, probability
and spectral analysis.
Each suite pairs a runnable rl_env/ with a preserved blind_run/ by
GPT-5.6 Sol Pro. The model name describes the evaluation actor… See the full description on the dataset page: https://huggingface.co/datasets/ulamai/Math-RL-Tasks.egms-qa-dataset
EGMS-QA Dataset
Prepared EGMS displacement tiles, encoder tokens, task labels, reference tables,
and natural-language QA records for 10,000 overlapping 7 km tiles. This card
describes the available data, file formats, and download options.
Data access
Data needed
Files to download
Details
Published QA records
train.jsonl, validation.jsonl, test.jsonl
QA loading example
Encoder inputs
Source tiles, metadata
Encoder data
Translator inputs
Token cache… See the full description on the dataset page: https://huggingface.co/datasets/risenyard/egms-qa-dataset.GLM-5.1-Reasoning-1M-Cleaned
GLM-5.1-Reasoning-1M-Cleaned
GLM-5.1-Reasoning-1M-Cleaned is a cleaned and reformatted derivative of Kassadin88/GLM-5.1-1000000x. It preserves the original four-subset layout (main, PHD-Science, Multilingual-STEM, Math) while converting every example into a unified SFT-ready schema with explicit conversations, input, output, domain, and meta fields.
This release was prepared from the original dataset published by Kassadin88.
Summary
Teacher model in the data: GLM-5.1… See the full description on the dataset page: https://huggingface.co/datasets/Jackrong/GLM-5.1-Reasoning-1M-Cleaned.math-code-science-deepseek-r1-en
R1 Dataset Collection
Aggregated high-quality English prompts and model-generated responses from DeepSeek R1 and DeepSeek R1-0528.
Dataset Summary
The R1 Dataset Collection combines multiple public DeepSeek-generated instruction-response corpora into a single, cleaned, English-only JSONL file. Each example consists of a <|user|> prompt and a <|assistant|> response in one "text" field. This release includes:
~21,000 examples from the DeepSeek-R1-0528 Distilled Custom… See the full description on the dataset page: https://huggingface.co/datasets/Hugodonotexit/math-code-science-deepseek-r1-en.claude-opus-4.6-4.7-reasoning-8.7k
Background
Ended up with some tokens to burn on a Claude Max plan. Assembly began during 4.6 and moved to 4.7. Model is tagged. The development evolved as it went along. The dataset has not been manually reviewed. It's entirely Claude developed.
Clarification on Reasoning
The reasoning is not Claude's actual chain-of-thought (cot) and is not summarized cot. It's a fully synthetic cot created as part of the Assistant response to mimic the type of "thinking"… See the full description on the dataset page: https://huggingface.co/datasets/angrygiraffe/claude-opus-4.6-4.7-reasoning-8.7k.tofu_resplitRAMDocs
RAMDocs
Data for the paper Retrieval-Augmented Generation with Conflicting Evidence.
RAMDocs is a dataset that simulates complex and realistic scenarios for conflicting evidence for a user query, including ambiguity, misinformation, and noise. We provide the raw data file RAMDocs_test.jsonl.
Data Fields
Each instance contains the following fields:
question: The question
documents: list of documents, where each document contains the following fields:
text: text of the… See the full description on the dataset page: https://huggingface.co/datasets/HanNight/RAMDocs.Medical-R1-Distill-Data
Introduction
This dataset is an SFT dataset distilled from Deepseek-R1 (Full Power Version), based on medical verifiable problems from HuatuoGPT-o1.
The Chinese version of the dataset is available at FreedomIntelligence/Medical-R1-Distill-Data-Chinese.
The distillation originates from the native Deepseek-R1 API requests. We hope this distilled dataset can help initialize your models with the reasoning chain from R1. You can also use our previously built medical verified long… See the full description on the dataset page: https://huggingface.co/datasets/FreedomIntelligence/Medical-R1-Distill-Data.big-finance-benchmark
BigFinanceBench Public Release
arXiv | Website | GitHub | Blog post
Finance answers are only useful when another analyst can audit how they were produced. BigFinanceBench evaluates that full workflow: agents must produce a numerical answer, and their traces are graded against point-weighted rubrics for source choice, period, accounting definition, assumptions, adjustments, and calculation.
This release contains a 50-question stratified subset of the 928-item BigFinanceBench… See the full description on the dataset page: https://huggingface.co/datasets/RogoAI/big-finance-benchmark.OlymMATH
Challenging the Boundaries of Reasoning: An Olympiad-Level Math Benchmark for Large Language Models
This is the official huggingface repository for Challenging the Boundaries of Reasoning: An Olympiad-Level Math Benchmark for Large Language Models by Haoxiang Sun, Yingqian Min, Zhipeng Chen, Wayne Xin Zhao, Zheng Liu, Zhongyuan Wang, Lei Fang, and Ji-Rong Wen.
We have also released the OlymMATH-eval dataset on HuggingFace 🤗, together with a data visualization tool OlymMATH-demo… See the full description on the dataset page: https://huggingface.co/datasets/RUC-AIBOX/OlymMATH.Chinese-DeepSeek-R1-Distill-data-110k
中文基于满血DeepSeek-R1蒸馏数据集（Chinese-Data-Distill-From-R1）
🤗 Hugging Face | 🤖 ModelScope | 🚀 Github | 📑 Blog
注意：提供了直接SFT使用的版本，点击下载。将数据中的思考和答案整合成output字段，大部分SFT代码框架均可直接直接加载训练。
本数据集为中文开源蒸馏满血R1的数据集，数据集中不仅包含math数据，还包括大量的通用类型数据，总数量为110K。
为什么开源这个数据？
R1的效果十分强大，并且基于R1蒸馏数据SFT的小模型也展现出了强大的效果，但检索发现，大部分开源的R1蒸馏数据集均为英文数据集。 同时，R1的报告中展示，蒸馏模型中同时也使用了部分通用场景数据集。
为了帮助大家更好地复现R1蒸馏模型的效果，特此开源中文数据集。
该中文数据集中的数据分布如下：… See the full description on the dataset page: https://huggingface.co/datasets/Congliu/Chinese-DeepSeek-R1-Distill-data-110k.AssetOpsBench
AssetOpsBench
AssetOpsBench is a specialized benchmark designed for evaluating Large Language Models (LLMs) and Multi-Agent systems in industrial operations. It focuses on the intersection of sensor data interpretation, maintenance logic, and Prognostics and Health Management (PHM).
The benchmark enables researchers to test how effectively AI agents can manage complex industrial assets, such as compressors and hydraulic pumps, by applying rule-based logic and diagnostic… See the full description on the dataset page: https://huggingface.co/datasets/ibm-research/AssetOpsBench.legal-rag-bench
Legal RAG Bench ⚖️
Legal RAG Bench by Isaacus is a reasoning-intensive benchmark for assessing the end-to-end, real-world performance of production-grade legal RAG systems.
Legal RAG Bench is composed of 4,876 passages sampled from the Judicial College of Victoria’s Criminal Charge Book alongside 100 complex, handwritten questions demanding expert-level knowledge of Victorian criminal law and procedure to be answered correctly.
Legal RAG Bench is the first open dataset for the… See the full description on the dataset page: https://huggingface.co/datasets/isaacus/legal-rag-bench.