Skip to content

Datasets

Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.

Data type

For which models

Find a dataset for your goal

Answered by Gemini 4 · the chat is kept only in this tab

DatasetJetBrains/KStackTop pick

KStack Kotlin code collection

Largest collection of permissive-licensed Kotlin code from GitHub.

  • Text generation
  • Code
1K 151M–10M rows
Datasetbertin-project/alpaca-spanish

Spanish translation of Stanford Alpaca instructions

Between 10 and 100 thousand Spanish instruction-response pairs translated using GPT-3.5-turbo for language model fine-tuning.

  • Text generation
  • Spanish
1K 3810K–100K rows
Datasetcommunity-datasets/tapaco

TaPaCo multilingual paraphrase corpus in 73 languages

Between 1 and 10 million semantically equivalent sentences from Tatoeba for 73 languages.

  • Translation
  • Text classification
  • Afrikaans
  • Arabic
  • Azerbaijani
  • +9
1K 481M–10M rows
Datasettaesiri/arxiv_qa

arXiv papers question and answer dataset

English question-answer pairs based on arXiv scientific papers for training and evaluating QA systems.

  • Question answering
  • English
1K 139100K–1M rows
Datasettandevllc/offsec_redteam_codesTop pick

Code corpus for pentesting and offensive security

A massive corpus of ~30 billion tokens from GitHub repositories on pentesting, OSINT, vulnerability exploitation, and forensics.

  • Text generation
  • Fill mask
  • Text retrieval
  • English
1K 241M–10M rows🔒 on request
Datasettencent/CL-bench

Benchmark for checking in-context learning

1,899 tasks where the model must learn new rules and procedures from context rather than pre-training.

  • Text generation
  • English
1K 1491K–10K rows
DatasetTheFinAI/Fino1_Reasoning_Path_FinQA

Reasoning chains for financial questions (FinQA)

FinQA questions with step-by-step GPT-4o reasoning paths for training financial reasoning models.

  • Question answering
  • Text generation
  • English
1K 421K–10K rows🔒 on request
Datasetallenai/XYBench-datasetTop pick

LLM benchmark for questions with false premises

An English dataset of questions with erroneous assumptions to evaluate LLM pragmatic responses.

  • Question answering
  • English
998 310K–100K rows🔒 on request
Datasetmirobody/MedTEMPO-Bench

Medication reconciliation benchmark for LLMs

A longitudinal benchmark for verifying medication regimen consistency during hospital admissions.

  • Question answering
  • English
  • Chinese
992 181K–10K rows
Datasetmaveriq/bigbenchhard

Challenging BIG-Bench Hard benchmark for language models

A set of 1k–10k challenging English text tasks where LLMs historically lagged behind humans.

  • Question answering
  • Token classification
  • Text classification
  • English
987 471K–10K rows
DatasetVLR-CVC/DocVQA-2026Top pick

DocVQA 2026 document benchmark

A small document QA benchmark across eight domains for multimodal reasoning.

  • Visual question answering
  • Document question answering
  • Image text → text
  • English
976 74up to 1K rows
Datasetprometheus-eval/Feedback-Collection

Long answer evaluation feedback dataset

English feedback texts to train open LLMs in fine-grained long answer evaluation.

  • Text generation
  • Text classification
  • English
971 12110K–100K rows
Datasetxtools-at/Claude-Haiku-5.5-High-Reasoning-TraceInversion-3000x

Synthetic chains of thought via trace inversion

Up to 10k complex English tasks with step-by-step reasoning generated using trace inversion.

  • Text generation
  • English
956 71K–10K rows
DatasetSp1786/multiclass-sentiment-analysis-dataset

English Multiclass Sentiment Analysis Dataset

An English tabular and text corpus for sentiment classification and translation tasks.

  • Text classification
  • Translation
  • English
952 3210K–100K rows
Datasettahrirchi/uz-books-v2

Uzbek books v2 corpus (OCR)

Bilingual corpus of about 40,000 books in Uzbek across Latin and Cyrillic versions.

  • Text generation
  • Fill mask
  • Uzbek
944 710K–100K rows
DatasetalphaXiv/AgentAlphaXivTop pick

Autonomous AI agent logs and research

A collection of research papers, code, chats, and traces from autonomous scientific agents on a preprint server.

  • Text generation
  • Summarization
  • Text classification
  • English
934 4510K–100K rows
DatasetGD-ML/TransitLM

Public transit route planning

Data for training language models in public transit route planning across Chinese cities.

  • Text generation
  • Chinese
919 84100K–1M rows
Datasetllm-jp/magpie-sft-v1.0

Magpie Japanese Synthetic SFT Dataset

100k to 1M Japanese instruction-response pairs for SFT generated by Calm3 and Qwen2.5 models.

  • Text generation
  • Japanese
915 23100K–1M rows