datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
super_glue
Dataset Card for "super_glue"
Dataset Summary
SuperGLUE (https://super.gluebenchmark.com/) is a new benchmark styled after
GLUE with a new set of more difficult language understanding tasks, improved
resources, and a new public leaderboard.
Supported Tasks and Leaderboards
More Information Needed
Languages
More Information Needed
Dataset Structure
Data Instances
axb
Size of downloaded dataset files: 0.03 MB
Size of… See the full description on the dataset page: https://huggingface.co/datasets/aps/super_glue.openbookqa
Dataset Card for OpenBookQA
Dataset Summary
OpenBookQA aims to promote research in advanced question-answering, probing a deeper understanding of both the topic
(with salient facts summarized as an open book, also provided with the dataset) and the language it is expressed in. In
particular, it contains questions that require multi-step reasoning, use of additional common and commonsense knowledge,
and rich text comprehension.
OpenBookQA is a new kind of… See the full description on the dataset page: https://huggingface.co/datasets/allenai/openbookqa.MMLU-Pro
MMLU-Pro Dataset
MMLU-Pro dataset is a more robust and challenging massive multi-task understanding dataset tailored to more rigorously benchmark large language models' capabilities. This dataset contains 12K complex questions across various disciplines.
|Github | 🏆Leaderboard | 📖Paper |
🚀 What's New
[2026.03.11] Added more cutting-edge frontier models to the leaderboard, including the Claude-4.6 series, Seed2.0 series, Qwen3.5 series, and Gemini-3.1-Pro… See the full description on the dataset page: https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro.codah
Dataset Card for COmmonsense Dataset Adversarially-authored by Humans
Dataset Summary
The COmmonsense Dataset Adversarially-authored by Humans (CODAH) is an evaluation set for commonsense
question-answering in the sentence completion style of SWAG. As opposed to other automatically generated
NLI datasets, CODAH is adversarially constructed by humans who can view feedback from a pre-trained model
and use this information to design challenging commonsense questions.… See the full description on the dataset page: https://huggingface.co/datasets/jaredfern/codah.distilabel-capybara-dpo-7k-binarized
Capybara-DPO 7K binarized
A DPO dataset built with distilabel atop the awesome LDJnr/Capybara
This is a preview version to collect feedback from the community. v2 will include the full base dataset and responses from more powerful models.
Why?
Multi-turn dialogue data is key to fine-tune capable chat models. Multi-turn preference data has been used by the most relevant RLHF works (Anthropic, Meta Llama2, etc.). Unfortunately, there are very few… See the full description on the dataset page: https://huggingface.co/datasets/argilla/distilabel-capybara-dpo-7k-binarized.stackoverflow-posts
StackOverflow Posts Markdown
Dataset Summary
This dataset contains all posts submitted to StackOverflow before the 14th of June 2023 formatted as Markdown text.
The dataset contains ~60 Million posts, totaling ~35GB in size and ~65 billion characters of text.
The data is sourced from Internet Archive StackExchange Data Dump.
Dataset Structure
Each record corresponds to one post of a particular type.
Original ordering from the data dump is not exactly preserved… See the full description on the dataset page: https://huggingface.co/datasets/mikex86/stackoverflow-posts.KodCode-V1-SFT-R1
🐱 KodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Coding
KodCode is the largest fully-synthetic open-source dataset providing verifiable solutions and tests for coding tasks. It contains 12 distinct subsets spanning various domains (from algorithmic to package-specific knowledge) and difficulty levels (from basic coding exercises to interview and competitive programming challenges). KodCode is designed for both supervised fine-tuning (SFT) and RL tuning.
🕸️… See the full description on the dataset page: https://huggingface.co/datasets/KodCode/KodCode-V1-SFT-R1.xcopa
Dataset Card for "xcopa"
Dataset Summary
XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning
The Cross-lingual Choice of Plausible Alternatives dataset is a benchmark to evaluate the ability of machine learning models to transfer commonsense reasoning across
languages. The dataset is the translation and reannotation of the English COPA (Roemmele et al. 2011) and covers 11 languages from 11 families and several areas around
the globe. The dataset is… See the full description on the dataset page: https://huggingface.co/datasets/cambridgeltl/xcopa.open-india-law
Open India Law
Open, structured Indian primary law - plus the scrapers that build it.
Every judgment of the Supreme Court of India and all 25 High Courts, the decisions of 15
tribunals and regulators, and Central, State and Union Territory legislation down to the
individual section. Normalized to one schema, exclusively from official government sources.
Volume
Period
Court judgments
12,848,644
1950 to 2025
Tribunal and regulator matters
813,168
1985 to 2026… See the full description on the dataset page: https://huggingface.co/datasets/vaquill/open-india-law.Legal_Corpus_QA_SynDeepThink
🧠 Legal Corpus QA SynDeepThink Dataset
This repository contains a high-intelligence Legal Question-and-Answer dataset, generated through an advanced Iterative and Recursive Thinking process. It bridges the gap between static legal corpora and the dynamic "check-and-recheck" nature of human legal expertise. 🏛️
💡 The Concept: Iterative & Recursive Legal Logic
While standard synthetic datasets are often generated in a single pass, Legal_Corpus_QA_SynDeepThink mimics the… See the full description on the dataset page: https://huggingface.co/datasets/Azzindani/Legal_Corpus_QA_SynDeepThink.PromptEval_MMLU_full
MMLU Multi-Prompt Evaluation Data
Overview
This dataset contains the results of a comprehensive evaluation of various Large Language Models (LLMs) using multiple prompt templates on the Massive Multitask Language Understanding (MMLU) benchmark. The data is introduced in
Maia Polo, Felipe, Ronald Xu, Lucas Weber, Mírian Silva, Onkar Bhardwaj, Leshem Choshen, Allysson Flavio Melo de Oliveira, Yuekai Sun, and Mikhail Yurochkin. "Efficient multi-prompt evaluation of LLMs."… See the full description on the dataset page: https://huggingface.co/datasets/PromptEval/PromptEval_MMLU_full.anle-toaan-gov-vn
Vietnamese Án lệ Corpus — anle.toaan.gov.vn
🇻🇳 Tóm tắt. Bộ dữ liệu các bản án + án lệ Việt Nam thu thập từ cổng
anle.toaan.gov.vn của Tòa án nhân dân tối cao.
Mỗi văn bản đi kèm markdown chuẩn hoá tiếng Việt và một lớp grounding mức câu
(mỗi trích dẫn mang sentence_id + char span trỏ ngược vào markdown). Bộ dữ
liệu là một phần của ViLA common-corpus và ship ba cấu hình HF theo chuẩn
chung: documents (bảng chính) · embeddings (vector 4096-D Nemotron-3-Embed-8B) ·
reduces (toạ… See the full description on the dataset page: https://huggingface.co/datasets/tmquan/anle-toaan-gov-vn.PromptEval_MMLU_correctness
MMLU Multi-Prompt Evaluation Data (correctness scores)
Overview
This dataset contains the results of a comprehensive evaluation of various Large Language Models (LLMs) using multiple prompt templates on the Massive Multitask Language Understanding (MMLU) benchmark. The data is introduced in
Maia Polo, Felipe, Ronald Xu, Lucas Weber, Mírian Silva, Onkar Bhardwaj, Leshem Choshen, Allysson Flavio Melo de Oliveira, Yuekai Sun, and Mikhail Yurochkin. "Efficient multi-prompt… See the full description on the dataset page: https://huggingface.co/datasets/PromptEval/PromptEval_MMLU_correctness.worldcup2026
⚽ WorldCup Arena
A Leakage-Free Forecasting Benchmark on a Live Tournament
Can a language model forecast a match — when the match had not been played at the moment it was asked?
🌐 Language / 语言 : 中文 ▾
📊 四张表
点开本页顶部的 Data Studio 标签即可浏览，也可以直接按名字加载。
Config
行数
内容
fixtures
104
基准本体 —— 喂给模型的头部信息，以及结算后的 90 分钟赛果，七个盘口全部推导好（outcome_1x2、over_2_5、both_score、odd_total）
dossiers
2,208
简报索引 —— 46 快照 × 48… See the full description on the dataset page: https://huggingface.co/datasets/Social-AI-2026/worldcup2026.Belle_1.4M-SLAM-Omni
Belle_1.4M
This dataset is prepared for the reproduction of SLAM-Omni.
This is a multi-round Chinese spoken dialogue training dataset. For code and usage examples, please refer to the related GitHub repository: X-LANCE/SLAM-LLM (examples/s2s)
🔧 Modifications
Data Filtering: We removed samples with excessively long data.
Speech Response Tokens: We used CosyVoice to synthesize corresponding semantic speech tokens for the speech response. These tokens, represented as… See the full description on the dataset page: https://huggingface.co/datasets/worstchan/Belle_1.4M-SLAM-Omni.AoPS-Scrape
AoPS-Scrape
Problems and solutions scraped from Art of Problem Solving (AoPS) Online class homework endpoints.
Obtained legally in accordance with AoPS's Terms of Service. This is not unauthorized redistribution of pirated material — access was through a legitimate authenticated AoPS Online class session.
Splits
Splits are named by scrape date (YYYY_MM_DD), plus a cross-date content-deduplicated split:
Split
Rows
Notes
deduplicated
29,964
One row per… See the full description on the dataset page: https://huggingface.co/datasets/hudsongouge/AoPS-Scrape.KodCode-Light-RL-10K
🐱 KodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Coding
KodCode is the largest fully-synthetic open-source dataset providing verifiable solutions and tests for coding tasks. It contains 12 distinct subsets spanning various domains (from algorithmic to package-specific knowledge) and difficulty levels (from basic coding exercises to interview and competitive programming challenges). KodCode is designed for both supervised fine-tuning (SFT) and RL tuning.
🕸️… See the full description on the dataset page: https://huggingface.co/datasets/KodCode/KodCode-Light-RL-10K.Omni-Frontier-Distillation-SFT-Cyber-Coding-Med-dataset-collection
🧬 Omni-Frontier Collection
Cybersecurity · Coding · Math · Science · RSI Reasoning — one unified SFT package
A unified, deduplicated, fully-browsable distillation & SFT corpus — every row real, every row visible.
📖 Jump to
What's inside · 🔁 Aggregation audit · 🛡 Cybersecurity · 💻 Coding · 🏭 Distillation deep-dive · 🔁 RSI · 🧮 Math/Science/More · 🎓 Training guide · 🔎 Browsing · 🧹 Quality · 🗺 Roadmap · 📄 License… See the full description on the dataset page: https://huggingface.co/datasets/SHSLab/Omni-Frontier-Distillation-SFT-Cyber-Coding-Med-dataset-collection.docmath-eval-failures-200
DocMath-Eval Failures 200: Agent Benchmark & Leaderboard
A curated benchmark of 200 challenging financial math questions that leading AI models
failed to answer correctly, with comprehensive evaluation results from multiple AI agents.
Leaderboard
Evaluated on 2026-02-21 using LLM-as-Judge (Qwen QwQ-32B) for soft scoring.
Rank
Agent
Model
Exact Match
Judge: Exact
Judge: Approx
Judge: Total
Wrong
Avg Duration
Avg Tool Calls
1
TRAE Agent
Opus 4.5
98/200 (49.0%)
96… See the full description on the dataset page: https://huggingface.co/datasets/Ayushnangia/docmath-eval-failures-200.ID_Legal_QA_SynDeepThink
🧠 Indonesian Legal QA SynDeepThink Dataset
This repository hosts a specialized Indonesian Legal QA dataset that incorporates a Deep Thinking Phase. It is engineered for researchers and developers focusing on high-level judicial reasoning and complex regulatory analysis. 🏛️
💡 The Concept: Deep Thinking vs. Standard QA
While standard models often provide "System 1" (snap) judgments, the SynDeepThink approach simulates "System 2" (slow, deliberate) thinking. This dataset… See the full description on the dataset page: https://huggingface.co/datasets/Azzindani/ID_Legal_QA_SynDeepThink.UltraChat-300K-SLAM-Omni
UltraChat-300K
This dataset is prepared for the reproduction of SLAM-Omni.
This is a multi-round English spoken dialogue training dataset. For code and usage examples, please refer to the related GitHub repository: X-LANCE/SLAM-LLM (examples/s2s)
🔧 Modifications
Data Filtering: We removed samples with excessively long data.
Speech Response Tokens: We used CosyVoice to synthesize corresponding semantic speech tokens for the speech response. These tokens, represented as… See the full description on the dataset page: https://huggingface.co/datasets/worstchan/UltraChat-300K-SLAM-Omni.CXM_Arena
Dataset Card for CXM Arena Benchmark Suite
Dataset Description
This dataset, "CXM Arena Benchmark Suite," is a comprehensive collection designed to evaluate various AI capabilities within the Customer Experience Management (CXM) domain. It consolidates five distinct tasks into a unified benchmark, enabling robust testing of models and pipelines in business contexts. The entire suite was synthetically generated using advanced large language models, primarily… See the full description on the dataset page: https://huggingface.co/datasets/sprinklr-huggingface/CXM_Arena.terminal-bench
Terminal-Bench Dataset
This dataset contains tasks from Terminal-Bench, a benchmark for evaluating AI agents in real terminal environments. Each task is packaged as a complete, self-contained archive that preserves the exact directory structure, binary files, Docker configurations, and test scripts needed for faithful reproduction.
The archive column contains a gzipped tarball of the entire task directory.
Dataset Overview
Terminal-Bench evaluates AI agents on… See the full description on the dataset page: https://huggingface.co/datasets/ia03/terminal-bench.reward-bench-2Code | Leaderboard | Results | Paper
RewardBench 2 Evaluation Dataset Card
The RewardBench 2 evaluation dataset is the new version of RewardBench that is based on unseen human data and designed to be substantially more difficult! RewardBench 2 evaluates capabilities of reward models over the following categories:
Factuality (NEW!): Tests the ability of RMs to detect hallucinations and other basic errors in completions.
Precise Instruction Following (NEW!): Tests the ability of RMs… See the full description on the dataset page: https://huggingface.co/datasets/allenai/reward-bench-2.code-alchemy
CodeAlchemy
CodeAlchemy is a synthetic code dataset (~976.6B tokens, ~162M rows) designed for training and evaluating code language models. It consists of 5 training subsets covering a range of code-related tasks, and 2 evaluation subsets. All files are Parquet with zstd compression with on-disk size ~873 GB. Raw source files are not included due to ownership considerations and must be manually fetched as instructed below.
Dataset Statistics
Config… See the full description on the dataset page: https://huggingface.co/datasets/open-alchemy/code-alchemy.jeb-rag
JEB-Bench
Charging the Gate Rent: Measured-Energy Accounting for Adaptive Retrieval-Augmented Generation
⚠️ Status: under construction. Phase 0 (measurement validation) and Phase 1
(index construction) are landing now. The oracle matrix (bench/oracle/) is
populated in Phase 2 and this card will be revised when it is complete. Do not
cite numbers from this repository until the status line says complete.
What this is
The first public per-query × per-configuration… See the full description on the dataset page: https://huggingface.co/datasets/Shanmuk4622/jeb-rag.KodCode-V1-SFT-4o
🐱 KodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Coding
KodCode is the largest fully-synthetic open-source dataset providing verifiable solutions and tests for coding tasks. It contains 12 distinct subsets spanning various domains (from algorithmic to package-specific knowledge) and difficulty levels (from basic coding exercises to interview and competitive programming challenges). KodCode is designed for both supervised fine-tuning (SFT) and RL tuning.
🕸️… See the full description on the dataset page: https://huggingface.co/datasets/KodCode/KodCode-V1-SFT-4o.SmallTalks
SmallTalks
Dataset description
SmallTalks is a synthetic dataset designed for supervised fine-tuning of language models. The dataset covers a variety of conversational content, including daily conversations, tool usage, Python programming, encyclopedia Q&A, exam problem-solving, logical reasoning, and more. Each task is provided in both English and Chinese versions.
You can load a dataset with the following command:
from datasets import… See the full description on the dataset page: https://huggingface.co/datasets/SmallDoge/SmallTalks.ledger-long-context-KPI-QA
LEDGER — Long-Context KPI Question Answering & Page Retrieval
This dataset is part of the LEDGER (Long-context Evaluation of Documents for
Grounded Extraction and Retrieval) benchmark.
It supports two of the three LEDGER tasks:
Page-level KPI retrieval — given a natural-language question about a financial
KPI and the corresponding annual report, retrieve the relevant page(s). Each row
includes TREC-style graded relevance judgments (qrels) over all candidate pages.… See the full description on the dataset page: https://huggingface.co/datasets/artefactory/ledger-long-context-KPI-QA.ID_Legal_QA_SynThink
🧠 Indonesian Legal QA Synthetic Think Dataset (ID_Legal_QA_SynThink)
This repository features an advanced Synthetic Question-and-Answer dataset for the Indonesian legal domain, distinguished by the inclusion of an explicit Thinking Phase (Chain-of-Thought). 🏛️
💡 The Concept: Transparent Legal Reasoning
Standard QA datasets often provide just the "final answer." This dataset goes deeper by capturing the internal reasoning process of the model before it arrives at a… See the full description on the dataset page: https://huggingface.co/datasets/Azzindani/ID_Legal_QA_SynThink.