datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
FineWeb2024
FineWeb-Edu 2024 — Cleaned and Shuffled
This dataset is a year-specific, cleaned, shuffled, and sharded release derived from HuggingFaceFW/fineweb-edu. It contains English educational web text collected in Common Crawl snapshots whose dump identifier belongs to 2024.
This is not a news-only dataset. The year refers to the Common Crawl capture year, not necessarily the page's publication year.
Dataset summary
Item
Value
Year
2024
Rows
162,500,784… See the full description on the dataset page: https://huggingface.co/datasets/Yahoo-Finance-News/FineWeb2024.Sujet-Finance-Instruct-177k
Sujet Finance Dataset Overview
The Sujet Finance dataset is a comprehensive collection designed for the fine-tuning of Language Learning Models (LLMs) for specialized tasks in the financial sector. It amalgamates data from 18 distinct datasets hosted on HuggingFace, resulting in a rich repository of 177,597 entries. These entries span across seven key financial LLM tasks, making Sujet Finance a versatile tool for developing and enhancing financial applications of AI.… See the full description on the dataset page: https://huggingface.co/datasets/sujet-ai/Sujet-Finance-Instruct-177k.FineWeb2025
FineWeb-Edu 2025 — Cleaned and Shuffled
This dataset is a year-specific, cleaned, shuffled, and sharded release derived from HuggingFaceFW/fineweb-edu. It contains English educational web text collected in Common Crawl snapshots whose dump identifier belongs to 2025.
This is not a news-only dataset. The year refers to the Common Crawl capture year, not necessarily the page's publication year.
Dataset summary
Item
Value
Year
2025
Rows
99,022,205… See the full description on the dataset page: https://huggingface.co/datasets/Yahoo-Finance-News/FineWeb2025.FineWeb-2023
FineWeb-Edu 2023 — Cleaned and Shuffled
This dataset is a year-specific, cleaned, shuffled, and sharded release derived from HuggingFaceFW/fineweb-edu. It contains English educational web text collected in Common Crawl snapshots whose dump identifier belongs to 2023.
This is not a news-only dataset. The year refers to the Common Crawl capture year, not necessarily the page's publication year.
Dataset summary
Item
Value
Year
2023
Rows
104,280,950… See the full description on the dataset page: https://huggingface.co/datasets/Yahoo-Finance-News/FineWeb-2023.lossbench-finance-v1
LossBench finance-v1
Severity-weighted expected-loss evaluation for agents that touch money. Three finance back-office domains, mechanical ground truth, and a contamination certificate. Models are ranked by what their mistakes cost, not by raw accuracy.
Overview
Task count
2400
Domains
reconciliation, payment_repair, settlement
License
cc-by-4.0
Tasks
Each task is an agentic back-office scenario with a deterministic seed, an… See the full description on the dataset page: https://huggingface.co/datasets/caiotheodoro/lossbench-finance-v1.ko-finance-asr-corrections
ko-finance-asr-corrections
Frequency-annotated Korean ASR confusion pairs from finance/stock YouTube.
210 pairs
mined from 2,391 videos of auto-captions
across 47 channels
totalling 1,080.1 hours
Each pair carries how often the term was mangled and how often it was said correctly, plus
verification provenance.
한국어 금융·주식 유튜브 자동자막에서 실측한 ASR 오인식→교정 쌍입니다. 모든 쌍에 오표기·정답
표기 빈도(→ 용어별 오인식률)와 검증 메타데이터(2-LLM 합의 감사, 승격 티어)가 붙어 있습니다.
What makes it different
No public… See the full description on the dataset page: https://huggingface.co/datasets/woongstar/ko-finance-asr-corrections.aprm-sft-thoughts-snorkel-finance-policy_best-adamw30-lp0
Act-PRM SFT thoughts — snorkel-finance finance
Act-PRM (Action Process Reward Models) infers the latent thoughts behind
logged, action-only agent demonstrations via an offline EM. For each
logged action x in state s we sample G=4 candidate thoughts z,
score each by the length-penalized action likelihood
reward(z) = p(x | s, z)
(len_frac grows with the thought's token length), and mark the best thought
(argmax reward). The (thought + action) span is then what downstream SFT / RL… See the full description on the dataset page: https://huggingface.co/datasets/mzio/aprm-sft-thoughts-snorkel-finance-policy_best-adamw30-lp0.finance-alpaca-1k-testindian-finance-synthetic-phase2-cleaned
Indian Finance Synthetic Dataset (Phase 2 - Final Clean)
Dataset Description
14,763 high-quality synthetic conversations about Indian personal finance, optimized for fine-tuning.
Recent Updates
✅ v3 (Final): Removed 14 samples with empty content messages
✅ v2: Removed 58 incomplete conversations
✅ v1: Tools optimization (82.5% size reduction)
All conversations are now complete and properly formatted for training.
Key Features
Clean… See the full description on the dataset page: https://huggingface.co/datasets/Gandalf1/indian-finance-synthetic-phase2-cleaned.Islamic_Finance_QnA_eval
Islamic Finance Q&A Evaluation Dataset
Validation and test splits for evaluating models on Islamic Finance Q&A.
Dataset Structure
Format: Simple prompt-answer pairs
Validation: ~203 examples (10%)
Test: ~203 examples (10%)
Language: Arabic
Domain: Islamic finance and Sharia-compliant banking
Fields
id: Unique identifier
prompt: The question prompt
question: Original question text
answer: Ground truth answer
topic: Topic category
split:… See the full description on the dataset page: https://huggingface.co/datasets/SahmBenchmark/Islamic_Finance_QnA_eval.Islamic_Finance_QnA_train
Islamic Finance Q&A Training Dataset
Training split of the Islamic Finance Q&A dataset in conversational format.
Dataset Structure
Format: Conversational (human-agent pairs)
Size: ~1,624 training examples (80% of total)
Language: Arabic
Domain: Islamic finance and Sharia-compliant banking
Usage
from datasets import load_dataset
dataset = load_dataset("SahmBenchmark/Islamic_Finance_QnA_train")
train_data = dataset['train']
# Example
example =… See the full description on the dataset page: https://huggingface.co/datasets/SahmBenchmark/Islamic_Finance_QnA_train.wdb-islamic-finance-benchmark
WDB Benchmark: Western Default Bias in Islamic Finance
Dataset Description
This benchmark tests whether Large Language Models exhibit Western Default Bias (WDB) - the tendency to provide Western/conventional finance answers even when the context implies Islamic finance should be used.
The Problem
When a user in Saudi Arabia or UAE asks a financial question, they likely expect Shariah-compliant advice. However, LLMs trained predominantly on Western data may… See the full description on the dataset page: https://huggingface.co/datasets/Raniahossam33/wdb-islamic-finance-benchmark.Finance-Instruct-AzerbaijaniThis is part of a translated version of the original dataset: https://huggingface.co/datasets/Josephgflowers/Finance-Instruct-500k
tw-finance-159M
Dataset Card for tw-finance-159M
本資料集收錄臺灣金融、財經、產業類新聞與專題文章之繁體中文文本，總 token 數約 159M（159 百萬），可作為繁中模型在「臺灣財經語境」下的補充預訓練語料。
Dataset Details
Dataset Description
資料來自繁體中文公開財經、產業類報導，內容涵蓋：
房地產／社會住宅政策
產業動態（科技、餐飲、農業、能源等）
個人理財、消費議題
商業活動／企業合作
每筆樣本以 text 為主文，搭配 token_count、word_count、url、updated_at 等 metadata，方便後續清理、去重與時間追蹤。
Curated by: Huang Liang Hsun
Language(s) (NLP): Traditional Chinese
License: cc-by-nc-sa-4.0
Dataset Sources… See the full description on the dataset page: https://huggingface.co/datasets/lianghsun/tw-finance-159M.spai-ss6-corpus-finance-business
SPAI SS6 Thai Finance Business Corpus Index
Index repo for the Thai finance/business corpus mirrored in the canonical repo.
This is a lightweight index dataset repo. It does not duplicate the full corpus.
The full Parquet data lives in the canonical repository config below.
Canonical Data
Canonical repo: SPAISS6F1/spai-ss6-llm-1b-thai-corpus
Canonical config: finance_business_pythainlp_thai_financial_dataset
Rows in canonical config: 502,942
Parquet size in… See the full description on the dataset page: https://huggingface.co/datasets/SPAISS6F1/spai-ss6-corpus-finance-business.sa-finance-reasoning-mix
tejeshbhalladhanyog/sa-finance-reasoning-mix
A merged reasoning-distillation dataset combining a custom multi-agent
financial-reasoning pipeline with a stratified slice of GLM-5.1's
general-domain reasoning data.
Composition
source / subset
rows
sa_pipeline_qwen_max (decomposition + cluster_generation + mapreduce_single)
153,923
main
80,000
PHD-Science
20,000
Multilingual-STEM
20,000
Math
15,000
total
288,923
Format
Each row is a… See the full description on the dataset page: https://huggingface.co/datasets/tejeshbhalladhanyog/sa-finance-reasoning-mix.