Common Crawl

Category

base_weights

Provider

commoncrawl.org

Type

dataset

Level

unknown

Topic

AI Training Data

Track

n/a

Section

Open Data Directory

Format

n/a

Status

publishable

Commercial

conditional

Featured

no

Fast start

no

Sequence

n/a

Priority

n/a

Primary source

training_data_stack

Sources

training_data_stack, website_existing

ID

7140efd5e93222fe

Stack Exchange Data Dump

archive.org

Access: open. Q&A dumps

arXiv Bulk Data Access

lib.ncsu.edu

Access: open. arXiv metadata/full text access options

OpenWebText2

github.com

Open web text

Common Voice

Mozilla

Large multilingual speech dataset project for ASR, speech research, and voice tooling.

Wikimedia Dumps

dumps.wikimedia.org

Strong encyclopedic backbone for general knowledge and factual style.

PhysioNet

physionet.org

Canonical source for ECG, ICU, waveform, and related biomedical datasets.

OpenNeuro

openneuro.org

Best open hub for MRI/EEG/MEG/iEEG style data.

MIT OpenCourseWare

MIT

Excellent lecture notes, exams, and videos across advanced technical topics.

UCI Machine Learning Repository

UCI

Classic and modern ML datasets that are ideal for education, benchmarking, and tabular experiments.

The Stack v2

Hugging Face

Top open code corpus; huge language coverage.

Open Images V7

storage.googleapis.com

Large supervised vision dataset with labels, boxes, masks, relations, narratives.

NASA Open Data Portal

NASA

NASA metadata portal for open datasets spanning science, aeronautics, Earth, and exploration.