File size: 15,376 Bytes
c7f01b5
 
 
b536799
2f79f80
c7f01b5
 
 
 
 
a2fcd39
 
 
c7f01b5
f03551e
 
 
 
 
f3d92a2
bd4b538
ce000ed
 
 
bd4b538
c7f01b5
 
f03551e
c7f01b5
a2fcd39
f3d92a2
a2fcd39
 
6b4bbe2
2f79f80
6b4bbe2
5b9f5a4
779c00c
 
5b9f5a4
 
779c00c
 
5b9f5a4
c7f01b5
5b9f5a4
c7f01b5
5b9f5a4
779c00c
5b9f5a4
196f3aa
31e0e37
 
 
5b9f5a4
b0b70b8
5b9f5a4
 
33b2959
779c00c
5b9f5a4
33b2959
31e0e37
 
5b9f5a4
 
33b2959
5b9f5a4
f03551e
5b9f5a4
 
f03551e
5b9f5a4
45a60ff
5b9f5a4
292f1c0
5b9f5a4
 
 
f03551e
5b9f5a4
 
 
 
c7f01b5
 
cc02077
779c00c
f3d92a2
b0b70b8
c7f01b5
 
5b9f5a4
b0b70b8
196f3aa
5b9f5a4
33b2959
f03551e
779c00c
f03551e
cc02077
43250a8
f03551e
 
c7f01b5
43250a8
cc02077
 
33b2959
5b9f5a4
513387e
f3d92a2
f03551e
5b9f5a4
779c00c
 
5b9f5a4
 
779c00c
 
5b9f5a4
 
 
a2fcd39
5b9f5a4
a2fcd39
 
24ad115
 
05a5218
 
 
37a0103
a2fcd39
05a5218
 
 
24ad115
acef550
5b9f5a4
a2fcd39
590d908
f03551e
5b9f5a4
 
a2fcd39
7e61ee8
5b9f5a4
 
 
 
779c00c
 
5b9f5a4
 
779c00c
 
5b9f5a4
 
a2fcd39
 
590d908
 
5b9f5a4
 
 
590d908
a6b0b9c
5b9f5a4
f03551e
a2fcd39
 
 
 
 
 
590d908
5b9f5a4
f3d92a2
 
 
2f79f80
 
 
779c00c
f3d92a2
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
---
license: cc0-1.0
size_categories:
- 1B<n<10B
pretty_name: Training Data Foundation
tags:
- data-catalog
- manifest
- urls
- provenance
- training-data
- pre-training
- post-training
configs:
- config_name: catalog
  default: true
  data_files:
  - split: train
    path:
    - meta/catalog.parquet
- config_name: preview
  data_files:
  - split: train
    path:
    - meta/preview.parquet
---
<!-- artifact-pointer-layout: v2 -->
<!-- Generated by artpool.index_projection from meta/, cards/*.yaml, the field registry and its templates. Every release regenerates it, so edits to this file do not last. -->

# Training Data Foundation

Real public data for pre-training, mid-training and post-training, each file with its address, size, hashes and license, fetched with one command;
this dataset is the foundation's public index.

English | [简体中文](README.zh.md) · [Data page](https://huggingface.co/spaces/Battam/training-data-foundation) · [Usage video](https://huggingface.co/spaces/Battam/training-data-foundation#usage-video)

<picture class="not-prose">
<source media="(max-width: 1023px)" srcset="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/stages-narrow-light.svg">
<img class="not-prose dark:hidden" src="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/stages-wide-light.svg" alt="What each training stage can use in this release. Pre-training (provisional): 30.7 TB deduplicated size, 151,161,846 rows, 30.0 TB allowed for training. Mid-training (provisional): 3.87 TB deduplicated size, 34,243,169 rows, 1.42 TB allowed for training. Post-training: 1,789 material sets, from 2,213 upstream sources, 1,406 allowed for training.">
</picture>
<picture class="not-prose">
<source media="(max-width: 1023px)" srcset="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/stages-narrow-dark.svg">
<img class="not-prose hidden dark:block" src="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/stages-wide-dark.svg" alt="What each training stage can use in this release. Pre-training (provisional): 30.7 TB deduplicated size, 151,161,846 rows, 30.0 TB allowed for training. Mid-training (provisional): 3.87 TB deduplicated size, 34,243,169 rows, 1.42 TB allowed for training. Post-training: 1,789 material sets, from 2,213 upstream sources, 1,406 allowed for training.">
</picture>

## Get data in one command

```sh
uv run https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/tools/fetch.py --kind material_set --form spreadsheets --out ./data
```

This fetches every spreadsheet material set whole into `./data`, one folder per unit, and checks each file against the size and hashes its row records.
Change `--kind` and `--form`, or add `--sample N` (at most N units of each source) or `--limit N` (N units in all), to take other material;
take a corpus, whose whole can be petabytes, with one of them;
the same command takes the same units of this release.

<details>
<summary>What the command takes, and in which order</summary>

The command and the queries below are pinned to the release tag `release-2026-11-17.3`.
[uv](https://docs.astral.sh/uv/) reads the script's dependencies, at the versions published before the day of this release.

The command returns units: every unit that passes the default profile of each material set whose form is `spreadsheets`, each in a folder of its own named after its `unit_id`.
`--sample N` takes the first N of each source (all of a source with fewer), in a fixed order: by the sha256 of the path of the data file that holds the unit, then by the unit's key;
`--limit N` takes N units of the whole selection from its sources in turn, each source's units in the same order: the first unit of each source, then the second of each, and so on, the sources of one turn in the order of the sha256 of their names.
The script reads the data files in that order (with `--limit`, the first file of each source, then the second of each) and stops once the files read hold the units, so the same command takes the same units of this release.

</details>

<details>
<summary>What the default profile leaves out</summary>

The default profile is the rules `tools/fetch.py` applies unless an option relaxes them.
It leaves out duplicate copies, rows that matched a benchmark or were not compared with the benchmark list yet, rows in a language other than English, files above 20 MB, and rows that the release's fetch check read and found refused by their site (HTTP 401 or 403, a login, or an empty or short body).
Each rule has its own option, and every run says how many units each rule removed.
The license and the personal-data mark are labels of every row, not rules of the default profile: every row comes whatever they say, and `--training-only` leaves out files whose license does not let them be shipped, `--no-personal-data` personal data.
Every rule and its option, and what to keep out of an agent's environment: [`docs/for-agents.md`](docs/for-agents.md).

</details>

<details>
<summary>Choose with SQL</summary>

Choose material sets from the catalog with [DuckDB](https://duckdb.org/):

```sql
SELECT source, collection, title, units, files, top_types, license.tier AS license_tier, needs_auth
FROM 'hf://datasets/Battam/training-data-foundation@release-2026-11-17.3/meta/catalog.parquet'
WHERE kind = 'material_set' AND needs_auth = 'no' AND license.tier IN ('permissive', 'share-alike')
ORDER BY collection, units DESC;
```

This returns material sets, one row each with its collection, counted in units (`units`);
with `kind = 'corpus'` it returns corpora instead, counted in rows (`files`).

Choose files in one corpus:

```sql
SET VARIABLE files = (SELECT list('hf://datasets/Battam/training-data-foundation@release-2026-11-17.3/' || f) FROM (SELECT unnest(data_files) AS f FROM 'hf://datasets/Battam/training-data-foundation@release-2026-11-17.3/meta/catalog.parquet' WHERE source = 'habs-haer-hals'));
SELECT uri, size_bytes, lower(hex(sha256)) AS sha256
FROM read_parquet(getvariable('files'))
WHERE source = 'habs-haer-hals' AND ext = 'pdf' AND size_bytes BETWEEN 100000 AND 5000000
ORDER BY size_bytes LIMIT 20;
```

This returns rows of the corpus `habs-haer-hals`, one per file.
The first statement takes its data files from the catalog's column `data_files`, so the query reads exactly those files and lists no folder of `data/`.
A data file may hold the rows of several sources, so the query keeps this one's by its column `source`.

</details>

## What's inside

<picture class="not-prose">
<source media="(max-width: 1023px)" srcset="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/inside-narrow-light.svg">
<img class="not-prose dark:hidden" src="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/inside-wide-light.svg" alt="The training stages by top-level domain, largest first. Pre-training (provisional), deduplicated size, by domain: Data science 29.9 TB; Science 531 GB; Machine learning 82.8 GB; Media 48.7 GB; Unclassified 47.0 GB; System administration 35.7 GB; Many domains 29.6 GB; Hardware 29.6 GB; Security 4.18 GB; Misc 376 MB; Software engineering 183 MB; Operations 4.08 KB. Mid-training (provisional), deduplicated size, by domain: System administration 1.27 TB; Many domains 1.00 TB; Machine learning 503 GB; Science 462 GB; Data science 429 GB; Hardware 177 GB; Security 13.9 GB; Unclassified 4.43 GB; Misc 3.45 GB; Software engineering 3.00 GB; Media 735 MB; Operations 13.9 MB. Post-training, material sets, by domain: Data science 353; Software engineering 286; Media 276; Operations 207; Machine learning 185; Science 179; Security 108; Hardware 87; System administration 71; Misc 37.">
</picture>
<picture class="not-prose">
<source media="(max-width: 1023px)" srcset="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/inside-narrow-dark.svg">
<img class="not-prose hidden dark:block" src="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/inside-wide-dark.svg" alt="The training stages by top-level domain, largest first. Pre-training (provisional), deduplicated size, by domain: Data science 29.9 TB; Science 531 GB; Machine learning 82.8 GB; Media 48.7 GB; Unclassified 47.0 GB; System administration 35.7 GB; Many domains 29.6 GB; Hardware 29.6 GB; Security 4.18 GB; Misc 376 MB; Software engineering 183 MB; Operations 4.08 KB. Mid-training (provisional), deduplicated size, by domain: System administration 1.27 TB; Many domains 1.00 TB; Machine learning 503 GB; Science 462 GB; Data science 429 GB; Hardware 177 GB; Security 13.9 GB; Unclassified 4.43 GB; Misc 3.45 GB; Software engineering 3.00 GB; Media 735 MB; Operations 13.9 MB. Post-training, material sets, by domain: Data science 353; Software engineering 286; Media 276; Operations 207; Machine learning 185; Science 179; Security 108; Hardware 87; System administration 71; Misc 37.">
</picture>

<details>
<summary>The training stages by top-level domain, as a table</summary>

| Domain | Pre-training: deduplicated size | Pre-training: rows | Mid-training: deduplicated size | Mid-training: rows | Post-training: material sets |
|---|---:|---:|---:|---:|---:|
| Software engineering | 183 MB | 59,085 | 3.00 GB | 256,404 | 286 |
| System administration | 35.7 GB | 1,999,552 | 1.27 TB | 11,396,233 | 71 |
| Data science | 29.9 TB | 107,371,726 | 429 GB | 270,472 | 353 |
| Machine learning | 82.8 GB | 237,332 | 503 GB | 230,420 | 185 |
| Science | 531 GB | 13,352,882 | 462 GB | 203,278 | 179 |
| Operations | 4.08 KB | 18,884,152 | 13.9 MB | 430 | 207 |
| Hardware | 29.6 GB | 388,557 | 177 GB | 4,945,550 | 87 |
| Security | 4.18 GB | 80,781 | 13.9 GB | 121,360 | 108 |
| Media | 48.7 GB | 16,401 | 735 MB | 35,708 | 276 |
| Misc | 376 MB | 2,403 | 3.45 GB | 102,989 | 37 |
| Many domains | 29.6 GB | 8,274,002 | 1.00 TB | 16,195,869 | 0 |
| Unclassified | 47.0 GB | 494,973 | 4.43 GB | 484,456 | 0 |

Each source is counted under its main domain, the top-level domain of its first domain code.
Every size is deduplicated, and every count is what a stage keeps after the rules every stage shares ([How it is made in the data page's manual](https://battam-training-data-foundation.static.hf.space/index.html?lng=en#manual/stage-counting)).

</details>

Browse by domain and training stage on the [data page](https://huggingface.co/spaces/Battam/training-data-foundation), and build a fetch command there.
Every material set and every corpus, with its counts, size, license tier and login, is on the pages of its domain, which [`docs/sources.md`](docs/sources.md) lists.

## How it is made

<picture class="not-prose">
<source media="(max-width: 1023px)" srcset="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/steps-narrow-light.svg">
<img class="not-prose dark:hidden" src="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/steps-wide-light.svg" alt="How the index is made, in four steps. 1, collect: collectors read whole public upstreams into corpora, one row per file or record, the material of pre-training and mid-training. 2, admit: works the search found upstream and groups of corpus rows become material sets, the material of post-training. 3, check: licenses, personal data, overlap with benchmarks and sizes are checked before release. 4, release: one commit with its own tag, release-2026-11-17.3.">
</picture>
<picture class="not-prose">
<source media="(max-width: 1023px)" srcset="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/steps-narrow-dark.svg">
<img class="not-prose hidden dark:block" src="https://huggingface.co/datasets/Battam/training-data-foundation/resolve/release-2026-11-17.3/docs/figures/steps-wide-dark.svg" alt="How the index is made, in four steps. 1, collect: collectors read whole public upstreams into corpora, one row per file or record, the material of pre-training and mid-training. 2, admit: works the search found upstream and groups of corpus rows become material sets, the material of post-training. 3, check: licenses, personal data, overlap with benchmarks and sizes are checked before release. 4, release: one commit with its own tag, release-2026-11-17.3.">
</picture>

Collectors read whole public upstreams into corpora, the material of pre-training and mid-training, and admission turns works found upstream and chosen corpus rows into material sets, the material of post-training.
Before each release every source is checked for its license, personal data, overlap with benchmarks and sizes, and each release is one tagged commit whose card and figures are computed from its own files.
Each step, and how sizes and the stage counts are measured: [How it is made in the data page's manual](https://battam-training-data-foundation.static.hf.space/index.html?lng=en#manual/made);
the glossary and the checks of this release: [Words and numbers in the data page's manual](https://battam-training-data-foundation.static.hf.space/index.html?lng=en#manual/words).

## Reference

Every page and file of this dataset, the fields used most, Croissant metadata, the Dataset Viewer and the SQL Console: [Pages and files of the dataset in the data page's manual](https://battam-training-data-foundation.static.hf.space/index.html?lng=en#manual/dataset).

## License, privacy and limits

Every row carries two labels: its file's training permit, `license_training_ok` (`yes`, `conditional`, `no` or `unknown`), and a personal-data mark when it has one: `health` or `minors` in its column `sensitive`, `personal_data` in its pool fields, or its source's `pii.level` in the catalog.
They are labels, not rules: `tools/fetch.py` takes every row whatever they say, and filters on them only when asked, `--training-only` leaving out files whose license does not let them be shipped and `--no-personal-data` rows marked as carrying personal data.
On the data page, the Training filter chooses rows by their permit, and each source's card says whether it holds personal data.
Rows that copy a public benchmark task, and rows not compared with the benchmark list yet, are left out of every training stage and of the fetch tool's default;
only `--allow-benchmark` and `--allow-not-compared` take them.
The catalog itself is released under CC0 1.0, and each file keeps the license of its source.
The full text, with the license details, personal information read from sample files, known limits, uses, releases and removal requests: [License, privacy and limits in the data page's manual](https://battam-training-data-foundation.static.hf.space/index.html?lng=en#manual/license).

## Citation

```bibtex
@misc{training_data_foundation,
  title        = {training-data-foundation: an index of public materials for building agent environments},
  howpublished = {https://huggingface.co/datasets/Battam/training-data-foundation},
  note         = {Release release-2026-11-17.3},
  year         = {2026}
}
```