Independent project. Not a U.S. government website.Independent project. Not a United States government website.
Support our projectSupport usSupport our project · Buy me a coffee (opens Buy Me a Coffee)Note: a newer version with 4.8 million conversations and demographic information can be found here. Dataset Description Paper: Interactive Search Tool: (paper) License:…
227 likes4,574 downloads in the last 30 daysText generationODC-By
This dataset has been created with distilabel. 📰 News [26/11/2024] 🆕 New version of the dataset is out! magpie-ultra-v1.0 is a new version of the MagPie Ultra dataset…
227 likes1,884 downloads in the last 30 daysText generationLlama 3.1 license
🌌 The Ultimate Open-Source Distillation Dataset — No Skip, Full, with Attack & Defense 🌌 Where 73 open-source minds converge into one unified stream of intelligence 18…
227 likes11,975 downloads in the last 30 daysText generationMIT
Dataset used to train GPT4All-J and GPT4All-J-LoRA We release several versions of datasets v1.0: The original dataset we used to finetune GPT-J on v1.1-breezy: A filtere…
226 likes467 downloads in the last 30 daysApache 2.0
This dataset contains 6.7 million cases from the Caselaw Access Project and Court Listener. The Caselaw Access Project consists of nearly 40 million pages of U.S. federa…
225 likes7,731 downloads in the last 30 daysText generationNo license given
This dataset was introduced in RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V Trustworthiness. GitHub This dataset was also used in MiniCPM-V 4.5: Cooking Effici…
224 likes1,157 downloads in the last 30 daysImage and text to textCC BY-NC 4.0
The Persuasion Dataset contains claims and corresponding human-written and model-generated arguments, along with persuasiveness scores. This dataset was created for rese…
222 likes1,197 downloads in the last 30 daysCC BY-NC-SA 4.0
This repository contains the data associated with the paper Goku: Flow Based Video Generative Foundation Models. Project page:
222 likes418 downloads in the last 30 daysText to videoNo license given
Using LLMs to improve other LLMs, at scale! OpenHermesPreferences is a dataset of ~1 million AI preferences derived from teknium/OpenHermes-2.5. It combines responses fr…
221 likes835 downloads in the last 30 daysText generationOther license
DeepPlanning: Benchmarking Long-Horizon Agentic Planning with Verifiable Constraints DeepPlanningBench is a challenging benchmark for evaluating long-horizon agentic pla…
221 likes3,619 downloads in the last 30 daysText generationApache 2.0
OpenCodeInterpreter: Integrating Code Generation with Execution and Refinement [🏠Homepage] | [🛠️Code] OpenCodeInterpreter OpenCodeInterpreter is a family of open-sourc…
220 likes24,598 downloads in the last 30 daysQuestion answeringApache 2.0
Coding agent session traces for badlogicgames/pi-mono This dataset contains redacted coding agent session traces collected while working on The traces were exported with…
219 likes2,179 downloads in the last 30 daysText generationOther license
The Omnilingual ASR Corpus is a collection of spontaneous speech recordings and their transcriptions for 348 under-served languages. The corpus was collected as part of…
216 likes5,589 downloads in the last 30 daysSpeech recognitionCC BY 4.0
MedTrinity-25M, a comprehensive, large-scale multimodal dataset for medicine, covering over 25 million images across 10 modalities, with multigranular annotations for mo…
215 likes701 downloads in the last 30 daysQuestion answeringNo license givenGated
Our training dataset consists of approximately 40,000 unique mathematics problem-answer pairs compiled from: AIME (American Invitational Mathematics Examination) problem…
213 likes37,931 downloads in the last 30 daysMIT
Pretraining Effectively on S2ORC! The peS2o dataset is a collection of ~40M creative open-access academic papers, cleaned, filtered, and formatted for pre-training of la…
213 likes26,716 downloads in the last 30 daysText generationODC-By
TriviaqQA is a reading comprehension dataset containing over 650K question-answer-evidence triples. TriviaqQA includes 95K question-answer pairs authored by trivia enthu…
213 likes72,286 downloads in the last 30 daysQuestion answeringUnknown license
APPS is a benchmark for Python code generation, it includes 10,000 problems, which range from having simple oneline solutions to being substantial algorithmic challenges…
212 likes20,445 downloads in the last 30 daysText generationMIT
APEX–Agents is a benchmark from Mercor for evaluating whether AI agents can execute long-horizon, cross-application professional services tasks. Tasks were created by in…
211 likes57,355 downloads in the last 30 daysCC BY 4.0Gated
ShareGPT4Video Captions 4.8M is a set of GPT4-Vision-powered multi-modal captions data of videos. It is constructed to enhance modality alignment and fine-grained visual…
211 likes8,834 downloads in the last 30 daysVisual question answeringCC BY-NC 4.0
Interactive Search Tool: WildChat paper: WildVis paper: Point of Contact: Yuntian Deng Dataset Summary WildChat-4.8M is a collection of 3,199,860 conversations between h…
210 likes18,126 downloads in the last 30 daysText generationODC-By
🤗 [LongWriter Dataset] • 💻 [Github Repo] • 📃 [LongWriter Paper] LongWriter-6k dataset contains 6,000 SFT data with ultra-long output ranging from 2k-32k words in leng…
210 likes2,091 downloads in the last 30 daysText generationApache 2.0
[01/31/2024] We update the OpenAI Moderation API results for ToxicChat (0124) based on their updated moderation model on on Jan 25, 2024.[01/28/2024] We release an offic…
209 likes11,972 downloads in the last 30 daysText classificationCC BY-NC 4.0
UltraData-Code is a complete implementation of the UltraData L0-L4 tiered data management framework. It covers four code data states from L0 through L3, with each level…
209 likes32,862 downloads in the last 30 daysText generationApache 2.0
The Open Super-large Crawled ALMAnaCH coRpus is a huge multilingual corpus obtained by language classification and filtering of the Common Crawl corpus using the goclass…
209 likes448 downloads in the last 30 daysText generationCC0 1.0Gated
InteriorGS: 3D Gaussian Splatting Dataset of Semantically Labeled Indoor Scenes A comprehensive indoor scene dataset featuring 3D Gaussian representations with semantic…
209 likes2,847 downloads in the last 30 daysInteriorgs terms of useGated
SuperGLUE ( is a new benchmark styled after GLUE with a new set of more difficult language understanding tasks, improved resources, and a new public leaderboard. Support…
208 likes315,034 downloads in the last 30 daysText classificationOther license
This repository is dedicated to prompts used to perform in-context learning with starcoder. As a matter of fact, the model is an autoregressive language model that is tr…
208 likes144 downloads in the last 30 daysApache 2.0
A large-scale multimodal dataset of 4,000+ hours of human interactions for AI research 🖼️ Blog 🌐 Website 🎮 Demo 📦 GitHub 📄 Paper Human communication involves a comp…
208 likes115,173 downloads in the last 30 daysCC BY-NC 4.0
Reasoning Corpus 5M · Within 5k sequence length About Dataset This dataset contains reasoning chains from major AI models, such as: DeepSeek-v4 (both Pro and Flash), Dee…
208 likes2,557 downloads in the last 30 daysText generationApache 2.0
Recap-DataComp-1B is a large-scale image-text dataset that has been recaptioned using an advanced LLaVA-1.5-LLaMA3-8B model to enhance the alignment and detail of textua…
208 likes17,707 downloads in the last 30 daysZero-shot classificationCC BY 4.0
AG is a collection of more than 1 million news articles. News articles have been gathered from more than 2000 news sources by ComeToMyHead in more than 1 year of activit…
207 likes121,648 downloads in the last 30 daysText classificationUnknown license
This dataset is used for the training of the LLaVA-Video model. We only allow the use of this dataset for academic research and education purpose. For OpenAI GPT-4 gener…
207 likes65,034 downloads in the last 30 daysVisual question answeringNo license given
The ToolScale dataset is a key component of the ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestration project. It provides synthetic environme…
207 likes2,668 downloads in the last 30 daysText generationOther license
A large-scale dataset of 48,478 screen recording videos (~12,300 hours) of professional software being used, sourced from the internet. All videos have been trimmed to r…
206 likes34,274 downloads in the last 30 daysVideo classificationCC BY 4.0
HellaSwag: Can a Machine Really Finish Your Sentence? is a new dataset for commonsense NLI. A paper was published at ACL2019. Supported Tasks and Leaderboards More Infor…
206 likes300,574 downloads in the last 30 daysNo license given
In the digital expanse, a Tree of Knowledge grows, Its branches of code and words intertwine in prose. Synthetic leaves shimmer, human insights compose, A binary symphon…
205 likes1,326 downloads in the last 30 daysText generationCC BY-SA 4.0
This is a processed version of the Flan V2 dataset. I'm not affiliated with the creators, I'm just releasing the files in an easier-to-access format after processing. Th…
205 likes1,444 downloads in the last 30 daysApache 2.0
This is the training data of WizardLM. News 🔥 🔥 🔥 [08/11/2023] We release WizardMath Models. 🔥 Our WizardMath-70B-V1.0 model slightly outperforms some closed-source…
205 likes1,886 downloads in the last 30 daysMIT
204 likes13,967 downloads in the last 30 daysNo license given
ToolACE is an automatic agentic pipeline designed to generate Accurate, Complex, and divErse tool-learning data. ToolACE leverages a novel self-evolution synthesis proce…
204 likes36,305 downloads in the last 30 daysText generationApache 2.0
203 likes2,907 downloads in the last 30 daysVisual question answeringMITGated
Industry Documents Library (IDL) is a document dataset filtered from UCSF documents library with 19 million pages kept as valid samples. Each document exists as a collec…
203 likes11,273 downloads in the last 30 daysImage to textIdl train
This dataset contains computationally intensive, self-contained, and unambiguous STEM reasoning problems across Physics, Mathematics, Biology, and Chemistry. Problems re…
203 likes126 downloads in the last 30 daysQuestion answeringMIT
This is the second iteration of the popular NuminaMath dataset, bringing high quality post-training data for approximately 900k competition-level math problems. Each sol…
202 likes84,134 downloads in the last 30 daysText generationApache 2.0
Bitext - Customer Service Tagged Training Dataset for LLM-based Virtual Assistants Overview This hybrid synthetic dataset is designed to be used to fine-tune Large Langu…
202 likes9,469 downloads in the last 30 daysQuestion answeringCDLA SHARING 1.0
AgentTrove is the largest open-source collection of agentic interaction traces to date, released by the OpenThoughts-Agent team. It contains 1,696,847 rows drawn from 21…
202 likes4,739 downloads in the last 30 daysText generationApache 2.0
Warning: this dataset contains data that may be offensive or harmful. The data are intended for research purposes, especially research that can make models less harmful.…
202 likes12,970 downloads in the last 30 daysText generationCC BY-NC 4.0
🍮 The WHOLE FLAN Collection! 🍮 Overview This repository includes the full dataset from the FLAN Collection, totalling ~300GB as parquets. Generated using the official…
201 likes29,297 downloads in the last 30 daysCC BY 4.0
Ultra-FineWeb-L1 is a large-scale English web corpus built from Common Crawl snapshots. Within UltraData's L0-L4 tiered data management framework, it serves as the L1 fi…
200 likes17,675 downloads in the last 30 daysText generationApache 2.0
The largest dataset of electronic schematics and PCB layouts on the internet, built as an engineering reference for schematic and PCB layout work. It's a self-growing, a…
199 likes12,009 downloads in the last 30 daysText generationCC BY 4.0
51 004 ShareGPT conversations that teach LLMs when, how and whether to call tools.Built with the Nous Research Atropos RL stack in Atropos using a custom MultiTurnToolCa…
199 likes5,041 downloads in the last 30 daysQuestion answeringApache 2.0
SWE-bench Pro is a benchmark of long-horizon software engineering tasks drawn from real pull requests in 11 open-source repositories (Go, Python, JavaScript, TypeScript)…
199 likes44,224 downloads in the last 30 daysNo license given
[2023/01/06] We release the commercial-use version of MathPile, namely MathPile_Commercial. [2023/01/06] We release the new version (v0.2, cleaner version) of MathPile.…
198 likes165 downloads in the last 30 daysCC BY-NC-SA 4.0Gated
This repository contains the dataset accompanying the paper Nemotron-Math: Efficient Long-Context Distillation of Mathematical Reasoning from Multi-Mode Supervision. Cod…
198 likes7,728 downloads in the last 30 daysText generationCC BY 4.0
LongBench is a comprehensive benchmark for multilingual and multi-task purposes, with the goal to fully measure and evaluate the ability of pre-trained language models t…
198 likes55,940 downloads in the last 30 daysQuestion answeringNo license given
Nemotron-Post-Training-Dataset-v1 Release This dataset is a compilation of SFT data that supports improvements of math, code, stem, general reasoning, and tool calling c…
197 likes16,308 downloads in the last 30 daysCC BY 4.0
UPDATE: The Internet Archive has requested that this dataset be deleted (see discussion #2) because they consider the IA's metadata too unreliable to determine whether a…
197 likes618 downloads in the last 30 daysText generationCC0 1.0
Self-Instruct is a dataset that contains 52k instructions, paired with 82K instance inputs and outputs. This instruction data can be used to conduct instruction-tuning f…
197 likes539 downloads in the last 30 daysApache 2.0
This is a streamable version of the Multilingual LibriSpeech (MLS) dataset. The data archives were restructured from the original ones from OpenSLR to make it easier to…
195 likes39,776 downloads in the last 30 daysSpeech recognitionCC BY 4.0
🎉 This work, introducing the AutoMathText dataset and the AutoDS method, has been accepted to The 63rd Annual Meeting of the Association for Computational Linguistics (…
195 likes6,411 downloads in the last 30 daysText generationCC BY-SA 4.0
Open-PerfectBlend is an open-source reproduction of the instruction dataset introduced in the paper "The Perfect Blend: Redefining RLHF with Mixture of Judges". It's a s…
195 likes5,837 downloads in the last 30 daysApache 2.0
Homepage: Repository: Paper: Dataset Description Dataset Summary The LegalBench project is an ongoing open science effort to collaboratively curate tasks for evaluating…
195 likes26,532 downloads in the last 30 daysText classificationCC BY 4.0
FLORES+ is an evaluation benchmark dataset for multilingual machine translation. Dataset Details Dataset Description FLORES+ is a multilingual machine translation benchm…
195 likes12,280 downloads in the last 30 daysText generationCC BY-SA 4.0Gated
Textbooks are all you need : A SciPhi Collection Dataset Description With LLMs, we can create a fully open-source Library of Alexandria. As a first attempt, we have gene…
195 likes414 downloads in the last 30 daysLlama 2 license
For more open-source datasets, models, and methodologies, please visit our GitHub repository. AM-DeepSeek-R1-Distilled-1.4M is a large-scale general reasoning task datas…
194 likes2,420 downloads in the last 30 daysText generationCC BY-NC 4.0
A decontaminated version of evol-codealpaca-v1. Decontamination is done in the same way as StarCoder (bigcode decontamination process).
193 likes29,778 downloads in the last 30 daysText generationApache 2.0
192 likes15,964 downloads in the last 30 daysText generationApache 2.0
If the video cannot be displayed in your environment, open it directly: assets/syndata-demo.mp4 1. Overview SynData is a next-generation large-scale real-world multimoda…
192 likes23,577 downloads in the last 30 daysCC BY 4.0
At 12.4 million image-caption pairs, PD12M is the largest public domain image-text dataset to date, with sufficient size to train foundation models while minimizing copy…
192 likes4,725 downloads in the last 30 daysCDLA PERMISSIVE 2.0
191 likes201 downloads in the last 30 daysText generationApache 2.0
More Information needed
191 likes20,786 downloads in the last 30 daysApache 2.0
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 datas…
190 likes16,963 downloads in the last 30 daysQuestion answeringApache 2.0
Sky-T1_data_17k.json: The 17k training data used to train Sky-T1-32B-Preview. The final data contains 5k coding data from APPs and TACO, and 10k math data from AIME, MAT…
190 likes2,015 downloads in the last 30 daysApache 2.0
The Nemotron-Agentic-Tool-Use-v1 dataset is designed to strengthen models’ capabilities as interactive, tool-using agents. It focuses on multi-turn conversations where l…
190 likes3,906 downloads in the last 30 daysCC BY 4.0
Spider is a large-scale complex and cross-domain semantic parsing and text-to-SQL dataset annotated by 11 Yale students. The goal of the Spider challenge is to develop n…
190 likes9,371 downloads in the last 30 daysCC BY-SA 4.0
This is a Cleaned Python Dataset Covering 25,000 Instructional Tasks Overview The dataset has 4 key features (fields): instruction, input, output, and text.It's a rich s…
189 likes7,566 downloads in the last 30 daysText classificationNo license given
AnswerCarefullyは日本語LLM 出力の安全性・適切性に特化したインストラクションデータセットです。 このデータセットは、英語の要注意回答を集めた Do-Not-Answer データセット の包括的なカテゴリ分類に基づき、人手で質問・回答ともに日本語サンプルを集めたオリジナルのデータセットです。 データセットの詳細については…
189 likes1,085 downloads in the last 30 daysAnswer carefully dataset touGated
Rare Disease, Real Kid: MVA Hackathon 2026 - Dataset Challenge Space: SageBio/rare-disease-real-kid-mva-hackathon-2026 Submission Period: 24 August 2026 – 24 October 202…
189 likes517 downloads in the last 30 daysCC BY 4.0Gated
SWE-bench Verified is a subset of 500 samples from the SWE-bench test set, which have been human-validated for quality. SWE-bench is a dataset that tests systems’ abilit…
189 likes129,605 downloads in the last 30 daysNo license given
This dataset contains self-contained, verifiable, and unambiguous STEM reasoning problems across Physics, Mathematics, Biology, and Chemistry. Each problem: Requires mul…
189 likes235 downloads in the last 30 daysQuestion answeringMIT
The 10,000-hour head-and-wrist line of EgoSuite-Open100K. Data Bucket · Collection · EgoDemo · EgoStandard · Project page Explore EgoSuite-Open100K ↗ Data location: EgoP…
188 likes672 downloads in the last 30 daysVideo classificationCommercial training no resale v1.0Gated
Hệ thống AI kết hợp để tạo personas tổng hợp dựa trên phân bố thực tế của Việt Nam A compound AI approach to personas grounded in real-world distributions Tổng quan (Ove…
188 likes12,665 downloads in the last 30 daysText generationCC BY 4.0
In response to authors demanding that LLMs stop using their works, here's a copy of The Pile with all copyrighted content removed.Please consider using this dataset to t…
187 likes69,127 downloads in the last 30 daysOther license
Retrieval-Augmented Generation (RAG) Dataset 12000 Retrieval-Augmented Generation (RAG) Dataset 12000 is an English dataset designed for RAG-optimized models, built by N…
187 likes1,216 downloads in the last 30 daysQuestion answeringApache 2.0
The Open Super-large Crawled Aggregated coRpus is a huge multilingual corpus obtained by language classification and filtering of the Common Crawl corpus using the Ungol…
187 likes519 downloads in the last 30 daysFill maskCC0 1.0Gated
2023/08/26 - Filtered results now only contain pure english instruction and removed any mentioned of trained by OAI response Median sequence length : 471 We employed a m…
187 likes5,840 downloads in the last 30 daysText generationApache 2.0
The Twitter Financial News dataset is an English-language dataset containing an annotated corpus of finance-related tweets. This dataset is used to classify finance-rela…
187 likes7,244 downloads in the last 30 daysText classificationMIT
本数据仓库是专为 LaTeX_OCR 及 LaTeX_OCR_PRO 制作的数据，来源于 以及 以及我们自己构建。 如果这个数据仓库有帮助到你的话，请点亮 ❤️like ++ 后续追加新的数据也会放在这个仓库 ~~ 原始数据仓库在github LinXueyuanStdio/Data-for-LaTeX_OCR. 数据集 本仓库有 5…
186 likes681 downloads in the last 30 daysImage to textApache 2.0
LLaVA Visual Instruct CC3M 595K Pretrain Dataset Card Dataset details Dataset type: LLaVA Visual Instruct CC3M Pretrain 595K is a subset of CC-3M dataset, filtered with…
186 likes905 downloads in the last 30 daysOther license
Textbook-like Dataset: A High-Quality Resource for Small Language Models The idea is simply inspired by the Textbooks Are All You Need II: phi-1.5 technical report paper…
186 likes301 downloads in the last 30 daysText generationApache 2.0Gated
Contains the data for the HealthBench eval. For the reference implementation of HealthBench, see OpenAI's simple-evals repo.
186 likes7,208 downloads in the last 30 daysMIT
Legal Intelligent systems in Chinese require a combination of various abilities, including legal text understanding and generation. To achieve this, we have constructed…
186 likes978 downloads in the last 30 daysApache 2.0
The dataset is a "distilabeled" version of the widely used dataset: Intel/orca_dpo_pairs. The original dataset has been used by 100s of open-source practitioners and mod…
185 likes16,655 downloads in the last 30 daysApache 2.0
UltraData-RL-2609 is the L3 refined data for reinforcement learning within UltraData's L0-L4 tiered data management framework. Built for the RL stage of MiniCPM5-2B post…
185 likes6,501 downloads in the last 30 daysText generationApache 2.0
Synthetic programming textbooks generated with GPT-3.5 and retrieval. Very high quality, aimed at being used in a phi replication. Currently 115M tokens. Covers many lan…
185 likes234 downloads in the last 30 daysNo license given
text-to-image-2M: A High-Quality, Diverse Text-to-Image Training Dataset Citation @article{zou2026advancing, title = {Advancing Aesthetic Image Generation via Compositio…
184 likes1,711 downloads in the last 30 daysText to imageMIT
Scientific papers datasets contains two sets of long and structured documents. The datasets are obtained from ArXiv and PubMed OpenAccess repositories. Both "arxiv" and…
183 likes2,955 downloads in the last 30 daysSummarizationUnknown license
Secure-code dataset: correction and audit notice (2026-10-06) The chosen and rejected fields are original model-generated preference labels. They are not verified secure…
183 likes1,559 downloads in the last 30 daysApache 2.0
We release the training dataset of ChatQA. It is built and derived from existing datasets: DROP, NarrativeQA, NewsQA, Quoref, ROPES, SQuAD1.1, SQuAD2.0, TAT-QA, a SFT da…
183 likes1,144 downloads in the last 30 daysOther license
Help keep USASI useful.
Optional. No USASI account required. Payment takes place on the linked provider’s website (Buy Me a Coffee).
About supporting this project