One row per nuclide Wikidata knows by proton AND neutron count: 3,444 isotopes, the deep table underneath the periodic table, rostered by the registry trick (anything stating both Z and N is a nuclide) and carrying the largest dynamic range of any column on the platform: half-lives normalized to seconds from hydrogen-4's 1.4 x 10^-22 to tellurium-130's 9 x 10^31, fifty-four orders of magnitude, with a humanized twin so 1.8 x 10^11 also reads 5.7 thousand years, which is carbon-14 doing the world's radiocarbon dating from row 40. Each nuclide carries A = Z + N (an identity every row passes by test), atomic mass in daltons (guarded: a mass more than 1.5 daltons from A is a graph typo and goes NA, which caught chlorine-38 stating 1.0 and americium-242 wearing americium-244's mass), natural abundance on the 286 that occur on Earth, spin and parity, and the decay story: mode, branching fraction and daughter from 2,926 decay statements. Stability is honest three ways: Y from the graph or abundance, N from a half-life, NA when the graph is silent, and tin's famous ten stable isotopes decompose correctly into 8 stable plus 2 carrying measured double-beta lower bounds past 10^25 seconds. Isomers are deduplicated with care, because QID string order once handed uranium-235 its metastable twin's 33-minute half-life instead of 704 million years. Element names join from the platform's own Periodic Table dataset and all 118 elements reconcile by test. Technetium and promethium show zero stable rows, as nature insists. Built with a 12-test acceptance suite.
3,444 rows in one CSV file, with 17 columns in total: Nuclide, Nuclide_ID, Element, Symbol, Z, N, A, Atomic_Mass, Stable, Half_Life_Seconds, Half_Life_Readable, Abundance_Percent, Spin, Parity, Decay_Modes, Decays_To, Nuclide_URL.
It is built by programmatically scraping wikidata.org, last pulled on 2026-09-25. Datasets are versioned; older versions stay downloadable.
Yes. Every CodeSights dataset is completely free as a CSV download; a free account is all it takes. Anyone can preview the data without signing in.
Yes. The exact Python scraper that built it is viewable on the dataset page by any signed-in member, so every number is reproducible.
Z joins the Periodic Table dataset's Atomic_Number; every element there has nuclides here, verified by test.
Automated scraping leaves room for error and the underlying sources change over time, so no version is guaranteed accurate or complete. If a number matters, verify it against the original source.