stack-v3-docs
| Language | Samples |
|---|---|
| Markdown | 199,003,312 |
| TeX | 5,139,532 |
| reStructuredText | 3,067,032 |
| Text | 2,752,971 |
| RMarkdown | 1,959,373 |
| Org | 1,043,577 |
| AsciiDoc | 840,637 |
| Roff | 570,452 |
| Wikitext | 455,210 |
| Roff Manpage | 406,048 |
| Textile | 64,706 |
| Typst | 43,944 |
| Pod | 40,833 |
| Vim Help File | 38,634 |
| Texinfo | 21,402 |
- Exact duplicate samples were removed after extraction.
Licensing information
These subsets are derived from The Stack v3 train dataset, which is released under the ODC-By 1.0 license. Individual source files may have different original licenses; their use must comply with those terms, including attribution where applicable. Repository provenance and per-file license metadata (repo_path, commit_id, license_type, and detected_licenses) are preserved in each sample. License detection may be incomplete or inaccurate.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support