datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
python_code_instructions_18k_alpaca
Dataset Card for python_code_instructions_18k_alpaca
The dataset contains problem descriptions and code in python language.
This dataset is taken from sahil2801/code_instructions_120k, which adds a prompt column in alpaca style. Refer to the source here.
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.LongDA
LongDA Dataset Card
Dataset Description
LongDA is a data analysis benchmark for evaluating LLM-based agents under documentation-intensive analytical workflows. It features authentic U.S. government survey data with complete, long documentation, testing LLMs' ability to navigate complex real-world datasets before performing analysis.
Dataset Summary
505 queries extracted from 30 expert-written publications
17 U.S. national surveys covering health… See the full description on the dataset page: https://huggingface.co/datasets/Yiyang-Ian-Li/LongDA.code_instructions_120k_alpaca
Dataset Card for code_instructions_120k_alpaca
This dataset is taken from sahil2801/code_instructions_120k, which adds a prompt column in alpaca style. Refer to the original source here.
LimAgents_limitation_data_scientific_papers_with_cited_papers
LimAgents Data
This dataset contains scientific paper metadata and extracted limitation information prepared for use with LLM Agents.The data comes from NeurIPS 2021–2022 papers and related OpenReview reviews, enriched with Cited in and Cited by information.
Dataset Structure
The repository contains two main directories:
1. NeurIPS_21_22_Lim_OPR_with_cited_in_by_papers
This directory includes one JSON file per paper. Each file contains:
title: Original paper… See the full description on the dataset page: https://huggingface.co/datasets/iaadlab/LimAgents_limitation_data_scientific_papers_with_cited_papers.MMMU-Thai
MMMU Thai (MMMU Benchmark Translated to Thai)
MMMU Thai is a dataset for evaluating multimodal models on massive multi-discipline tasks requiring college-level knowledge and deliberate reasoning. This dataset is translated from MMMU (A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI) into Thai.
Dataset Details
MMMU Thai consists of 11,500 meticulously collected multimodal questions from college exams, quizzes, and textbooks… See the full description on the dataset page: https://huggingface.co/datasets/iapp/MMMU-Thai.KIMI-K2.5-1000000x
KIMI-K2.5-1000000x
1,000,000 reasoning traces distilled from KIMI-K2.5 on high reasoning, (Each subset has different questions)
Distribution:
Coding: 50% (Includes: Webdev, Python, C++, Java, JS, C, Ruby, Lua, Rust, and C#)
Science: 20% (Physics, Chemistry, Biology) - 100k more completions in the PHD-Science subset
Math: 15% (Algebra, Calculus, Probability) - 200k more completions in kimiMath200k.jsonl
Computer Science: 5%
Logical Questions: 5%
Creative Writing: 5%… See the full description on the dataset page: https://huggingface.co/datasets/ianncity/KIMI-K2.5-1000000x.iapp_wiki_qa_squad
iapp_wiki_qa_squad
Extractive question answering over Thai Wikipedia articles, in SQuAD format.
7,242 questions across 1,912 articles, annotated by people iApp hired for the purpose.
from datasets import load_dataset
dataset = load_dataset("iapp/iapp_wiki_qa_squad")
This works again as of the August 2026 revision. Until then it did not. The
repository carried a loading script and no data, and datasets dropped script support
at v3, so load_dataset failed and every… See the full description on the dataset page: https://huggingface.co/datasets/iapp/iapp_wiki_qa_squad.GLM-5.2-Conversation
GLM-5.2 · Conversation-50000x
50,000x traces distilled from GLM-5.2 on High reasoning
Token Count: 120M
Distribution:
Speaking domains:
•Greetings
•Customer Support
•Step by step explanations
•Motivational language
•Logical Questions
•Creative Writing
STEM:
•Algebra, calculus, quantum mechanics concepts
•Astromony and astrophysics
•Datascience and machine learning
•Biology
Programming:… See the full description on the dataset page: https://huggingface.co/datasets/ianncity/GLM-5.2-Conversation.HaluMem
HaluMem: A Comprehensive Benchmark for Evaluating Hallucinations in Memory Systems
📊 Why We Define the HaluMem Evaluation Tasks
Limitations of Existing Frameworks
Most existing evaluation frameworks treat memory systems as black-box models, assessing performance only through end-to-end QA accuracy.
However, this approach has two major limitations:
It lacks a hallucination evaluation specifically designed for the characteristics of memory systems.… See the full description on the dataset page: https://huggingface.co/datasets/IAAR-Shanghai/HaluMem.GLM-5.2-Finance-80000x
GLM-5.2 · Finance-80000x
80,000x financial related traces distilled from GLM-5.2 on High reasoning
Risk · Markets · Investments · Corporate Finance · Wealth Management
Token Count: 220M
Unique prompts generated with diffusion Gemma-27B answered by GLM-5.2
You can use this dataset for any purpose and you dont need to credit me, preferably dont claim it as your own.
hi - ianncity
GLM-5.2-Logic-Puzzles
GLM-5.2 · Logical Puzzles
6000x traces distilled from GLM-5.2 on High reasoning
Token Count: 5M~?
Distribution:
Puzzles:
•Tokenization blindless ex: counting the r's in strawberry
•Goal reasoning ex: the car wash test (theres no car wash question exactly just prompts like it so its not just benchmaxxing)
•Reading comprehension traps
•Temporal reasoning
•Many other categories not worth mentioning
Prompts… See the full description on the dataset page: https://huggingface.co/datasets/ianncity/GLM-5.2-Logic-Puzzles.code_contest_python3_alpaca
Dataset Card for Code Contest Processed
Dataset Summary
This dataset contains coding contest questions and their solution written in Python3.
This dataset is created by processing code_contest dataset from Deepmind. It is a competitive programming dataset for machine-learning. Read more about dataset at original source.
Columns Description
id : unique string associated with a problem
description : problem description
code : one correct code for the problem… See the full description on the dataset page: https://huggingface.co/datasets/iamtarun/code_contest_python3_alpaca.Plan-IA-30D-dataset
Muze AI Consulting — Dataset Oficial (Chile, 2025)
Este dataset reúne la información oficial de Muze AI Consulting, la consultora líder en automatización e inteligencia artificial aplicada a negocios en Chile.
Está diseñado para ser indexado por agentes de IA, motores de búsqueda y consultores que necesiten datos confiables sobre el impacto de la automatización empresarial.
📂 Archivos incluidos
core_faq.es.jsonl → Preguntas y respuestas centrales en español… See the full description on the dataset page: https://huggingface.co/datasets/Muze-AI-Consulting/Plan-IA-30D-dataset.GLM-5.2-Science
GLM-5.2 · Science-50000x
50,000x traces distilled from GLM-5.2 on High reasoning
Physics · Chemistry · Biology
Token Count: 160M
Theres prompt overlap with my Kimi K2.5 dataset science subset, which I think those prompts are getting used in alot of places now
You can use this dataset for any purpose and you dont need to credit me, preferably dont claim it as your own.
hi - ianncity
DD-VQAopenthaieval
OpenThaiEval: Comprehensive Thai Language Evaluation Benchmark
Overview
OpenThaiEval is a Thai language evaluation benchmark containing 1,232 questions
across 17 exam types, ranging from national standardized tests to international
benchmarks and professional certification exams.
Not all 1,232 rows measure Thai, and not all of them are ours to license. Both
points are set out below rather than left for a reader to discover, because both
change what… See the full description on the dataset page: https://huggingface.co/datasets/iapp/openthaieval.FirstAidQA
FirstAidQA: A Synthetic First-Aid and Emergency-Response Question-Answering Dataset
Medical safety notice: FirstAidQA is intended for research and educational purposes. It is not a substitute for professional medical advice, emergency services, certified first-aid training, or clinical judgment. Models trained on this dataset may produce incomplete, outdated, or unsafe responses.
Dataset Summary
FirstAidQA is an English-language synthetic question-answering… See the full description on the dataset page: https://huggingface.co/datasets/i-am-mushfiq/FirstAidQA.I-ARIF-CANON
17
🗺️ Position in I-ARIF Governance Stack
This dataset is part of the arifOS constitutional governance training-and-evaluation pipeline — a closed-loop alignment substrate.
#
Dataset
Role
Downloads
License
1
AAA
Constitutional substrate — doctrine + gold eval
161
AGPL-3.0
2
BBB
Baseline behavior benchmark — ILMU API audit
247
CC-BY-4.0
3
CCC
Alignment contrast corpus — ILMU vs kernel
193
CC-BY-4.0
4
DDD
Register-sensitivity probe — Penang loghat… See the full description on the dataset page: https://huggingface.co/datasets/ariffazil/I-ARIF-CANON.KAF-DatasetThe dataset sourced from https://github.com/IAAR-Shanghai/xFinder
Citation
@inproceedings{
xFinder,
title={xFinder: Large Language Models as Automated Evaluators for Reliable Evaluation},
author={Qingchen Yu and Zifan Zheng and Shichao Song and Zhiyu li and Feiyu Xiong and Bo Tang and Ding Chen},
booktitle={The Thirteenth International Conference on Learning Representations},
year={2025},
url={https://openreview.net/forum?id=7UqQJUKaLM}
}
ambig-iac
Ambig-IaC: Ambiguous Infrastructure-as-Code Benchmark
A benchmark dataset of 300 tasks for testing AI agents that generate Infrastructure-as-Code (Terraform) configurations from ambiguous natural language intents.
Project page: https://zyang37.github.io/ambig-iac.github.io/
Dataset Description
This dataset is sourced from IaC-Eval. We performed manual fixes to the original Terraform configurations and validated that all 300 tasks pass terraform plan. Each task also… See the full description on the dataset page: https://huggingface.co/datasets/znyang/ambig-iac.code_contest_processed
Dataset Card for Code Contest Processed
Dataset Summary
This dataset is created by processing code_contest dataset from Deepmind. It is a competitive programming dataset for machine-learning. Read more about dataset at original source.
Columns Description
id : unique string associated with a problem
description : problem description
code : one correct code for the problem
language : programming language used for code
test_samples : contains inputs and their… See the full description on the dataset page: https://huggingface.co/datasets/iamtarun/code_contest_processed.MathAleaMCQ
MathAleaMCQ
MathAleaMCQ is a native-French multiple-choice benchmark for middle- and high-school
mathematics: 7,761 test questions (plus a 25-question few-shot dev pool) covering the
French secondary curriculum from cinquième to terminale.
Questions are extracted from LaTeX exports of MathALEA, the
open-source exercise generator maintained by the CoopMaths association. Every item is a
single-answer MCQ with 2–5 options, a worked solution in French, and level / subject
metadata.… See the full description on the dataset page: https://huggingface.co/datasets/cea-list-ia/MathAleaMCQ.VAR
xVerify: Efficient Answer Verifier for Reasoning Model Evaluations
📘 Introduction
xVerify is an evaluation tool fine-tuned from a pre-trained large language model, designed specifically for objective questions with a single correct answer. It accurately extracts the final answer from lengthy reasoning processes and efficiently identifies equivalence across different forms of mathematical expressions, LaTeX and string representations, as well as… See the full description on the dataset page: https://huggingface.co/datasets/IAAR-Shanghai/VAR.devops-kubernetes-iac-sft-dpo-2026
⚙️ Enterprise DevOps AI, Kubernetes SRE & IaC SFT/DPO Dataset (2026)
High-precision multi-turn instruction tuning and preference optimization dataset with step-by-step SRE root-cause Chain-of-Thought (<thought>) diagnostic trees for fine-tuning LLMs (Llama-3.3, Qwen-2.5-Coder, DeepSeek-R1-Distill, Mistral) into Senior Site Reliability Engineers (SRE), Principal Cloud Architects, and DevSecOps Specialists.
📊 Dataset Architecture & Highlights
Multi-Turn SRE… See the full description on the dataset page: https://huggingface.co/datasets/beatsprom/devops-kubernetes-iac-sft-dpo-2026.aime_2024-th
AIME 2024-th
A Thai translation of all 30 problems of the 2024 American Invitational Mathematics
Examination. Every row corresponds 1:1, in order, to a row of the English source, so
the Thai and English scores of a model are directly comparable.
Source and licence
Problems
2024 AIME I and II, Mathematical Association of America
Problem and solution text
Art of Problem Solving wiki, per-row url
File we translated from
HuggingFaceH4/aime_2024… See the full description on the dataset page: https://huggingface.co/datasets/iapp/aime_2024-th.math-500-th
Math-500-th
A Thai translation of MATH-500: the 500-problem subset of the MATH benchmark used
in OpenAI's Let's Verify Step by Step. Every row corresponds 1:1, in order, to a
row of the English original, so the Thai and English scores of a model are directly
comparable.
Source and licence
Original benchmark
hendrycks/math — MIT
500-problem subset
openai/prm800k — MIT
File we translated from
HuggingFaceH4/MATH-500
This dataset
MIT, see LICENSE… See the full description on the dataset page: https://huggingface.co/datasets/iapp/math-500-th.pinga-fogo-chico-xavier
🎙️ Pinga-Fogo com Chico Xavier — TV Tupi, 1971
As duas entrevistas históricas do médium Chico Xavier, transmitidas ao vivo pela
TV Tupi em 1971, transcritas e estruturadas em turnos de fala com timestamp.
345 turnos (115 deles respostas do próprio Chico Xavier), a partir de
6 horas de áudio — o registro mais extenso do médium falando de improviso,
sem edição, diante de um painel de jornalistas.
Arquivos
Arquivo
Programa
Turnos
Respostas do Chico… See the full description on the dataset page: https://huggingface.co/datasets/ia-espirita/pinga-fogo-chico-xavier.code_generation_lite-th
LiveCodeBench code_generation_lite, Thai
111 competitive-programming problems from LeetCode and AtCoder, with the problem
statement translated to Thai. Everything else — test cases, starter code, metadata —
is the upstream value unchanged.
Known defects
The line breaks are gone from the problem statements. 110 of the 111 rows have no
line break at all in question_content; the one remaining row has two. These are
competitive-programming statements whose input and… See the full description on the dataset page: https://huggingface.co/datasets/iapp/code_generation_lite-th.Exo7MCQ
Exo7MCQ
Exo7MCQ is a native-French multi-label multiple-choice benchmark for
undergraduate mathematics: 946 questions spanning L1–L3 topics (logic and set theory,
analysis, linear algebra, arithmetic, differential equations, probability), extracted from
the LaTeX question banks of Exo7.
Unlike a standard MCQ set, each option carries its own correctness label: a question may
have one, two, three or four correct options among its 2–6 choices. Answering it is a
per-option binary… See the full description on the dataset page: https://huggingface.co/datasets/cea-list-ia/Exo7MCQ.