SilverPrice.org spot silver prices

One commodity, one obsession, fifty years of receipts. SilverPrice.org is the silver-facing site operated by the GoldPrice.org team, and its quote boxes hydrate from a structured feed that Datadory delivers as rows: spot silver (XAG) priced per troy ounce in dozens of national currencies, one live point per metal-currency pair, backed by a daily history of roughly 13,600 observations per pair running from January 1975 to the present.

A worked anchor shows the texture. During verification, the dollar series resolved to a live point of 69.3362 US dollars per troy ounce, sitting above recent daily closes of 63.97 (18 August 2026), 66.05 (17 August) and 64.93 (14 August) - a week that moved silver several percent, recorded in rows rather than screenshots. Chart views on the source side run from 30-day snapshots to the all-data series; the underlying rows are what ship to you.

Where this record earns its place in the Datadory catalog - 1,744 datasets across 159 viable industries - is longevity per column. Most metals feeds trade depth for breadth or the reverse; this one holds a single basis (the troy ounce), a single frequency (daily), and reaches back further than almost anything else in the precious metals slice. Get a sample of this dataset and the rows come back shaped exactly like the dictionary below, cut to the currency pairs you name.

One real record from the silver-on-dollars series, flat exactly as its fields arrive:

# one record from the silver-on-dollars series, as it lands
instrument      : USD-XAG
price           : 69.3362        # US dollars per troy ounce
timestamp       : 17870256       # native encoding; decodes to 2026-08-18

# the same row shape, walking back through August 2026
date            : 2026-08-18     close 63.97 USD/oz
date            : 2026-08-17     close 66.05 USD/oz
date            : 2026-08-14     close 64.93 USD/oz

# the anatomy every row repeats
basis           : one troy ounce of silver (XAG)
currencies      : USD | EUR | GBP | CAD | AUD | INR | CNY | ... dozens more
series          : ~13,600 daily observations per metal-currency pair, from January 1975

Read the anatomy rather than the numbers. First, shape: three typed columns, one row per observation, no nesting - the series lands in a dataframe without a parsing pass. Second, keys: the instrument field (USD-XAG) names both halves of the pair, which makes multi-currency work a filter and cross-metal joins a string swap. Third, time: the timestamp arrives in a compact native encoding that decodes to a plain calendar date, so ordering survives any timezone argument. Fourth, units: everything is one troy ounce, always, so nothing in the series ever needs normalizing before it can be compared.

Three documented fields, definitions verified during research - deliberately small, and sufficient. An identification column (instrument, the currency-metal pair), a value column (price, US dollars or the chosen currency per troy ounce), and a time column (timestamp, in the source's compact native encoding).

There is no hidden tail behind these three, and that is the point: the depth of this record lives in its rows (roughly 13,600 daily observations per pair) and its breadth (dozens of currency pairs), not in column count. What can be added on request rides on the same schema: decoded calendar dates, gram and kilogram unit variants matching the source's per-gram and per-kilo quoting, and cross-pairs such as USD-XAU (gold, for ratio work) and EUR-XAG (euro silver, for local-currency reads). Their shapes get confirmed against live rows when your sample is cut - which is also where column naming locks down for your pipeline.

Geography - a single world silver price expressed in the customer's currency rather than a venue-by-venue patchwork. Dozens of national currency pairs are published, with USD, EUR, GBP, CAD, AUD, INR and CNY among the named ones; the currency choice converts an FX question into a lookup.

Temporal - the series starts in January 1975, the first year of freely traded post-Bretton-Woods silver, and runs through the present. Delivered windows are selectable: a 30-day snapshot when the job is current-state, the full-history series when the job is regime analysis. Five decades means every fiat-era silver cycle - the 1980 corner, the 2011 spike, the 2020 squeeze - fits inside one panel.

Granularity - two scales on one basis. One live spot point per metal-currency pair carries the present market level; the full-history series carries daily observations, roughly 13,600 per pair. Both are denominated per troy ounce, which keeps aggregates reproducible: no pre-averaged indices, every downstream number computes from the atomic row.

Set against the wider catalog - 1,744 datasets averaging 7.81/10 - this record scores 6/10, and the reason is stated plainly: the three-field dictionary is the thinnest in the slice. The score buys depth of history, not width of schema. If the job needs lifecycle flags or fundamentals attached to each price, the industry hub routes to richer records; if the job is a clean, long, multi-currency silver series, this is the efficient shape.

API, files, or your warehouse. Daily, weekly, or hourly.

Name the currency pairs when you request the sample - the dollar series alone, a regional basket, or the full multi-currency grid, with or without the gold pair for ratio work. The sample ships first either way, shaped exactly like the dictionary above; the ongoing feed lands on whatever cadence your models need, into whatever channel your team already works in. Cadence changes are a settings conversation, not a re-integration project.

Five jobs a fifty-year, multi-currency silver series settles outright:

1. Gold-silver ratio work. The dollar silver series aligns with the dollar gold series on the same schema, so the ratio becomes one division over matched dates - and its five-decade history becomes a chart nobody has to reconstruct.
2. Multi-currency pricing displays. Dealers and finance products serving customers in rupees, yuan or francs quote silver in the viewer's own currency from the published pairs, retiring a hand-rolled FX layer.
3. Long-run monetary research. From 1975 onward, every fiat-era episode is inside the panel - inflation regimes, squeezes, drawdowns - at daily resolution.
4. Backtesting and feature engineering. One basis, one frequency, regularly spaced: the series drops into a pipeline as the silver leg under any metals strategy with zero cleaning passes.
5. Industrial demand modeling. Solar, electronics and brazing analysts need one consistent price input; a single world spot series beats stitching vendor quotes across regions.

For contrast within the same slice: LBMA Precious Metals Prices sets the institutional benchmarks, BullionVault Gold Price Chart & Historical Data supplies the intraday bars, and this record owns the deep, simple daily spine.

Investors & Quant Researchers get the deepest fit: five decades of daily silver in any major currency, with the ratio leg one join away. Data Scientists & ML Engineers get the smallest plausible schema - three typed columns - ready as targets or features without a labeling pass. Developers & Builders get a pair-keyed feed where picking a currency is picking a filter value, which keeps integrations measured in hours. Market Researchers & Consultants get local-currency evidence for client decks across dozens of markets. Competitive Intelligence & Product Teams at bullion retailers get a neutral world spot baseline to audit their own displayed prices against. Persona-by-persona detail lives on the competitive intel, data science and developer industry pages, and the shortlist comparison sits on best precious metals & minerals datasets.

Provenance note - published by Silverprice.org, the silver property of the GoldPrice.org operation, whose quote infrastructure serves both metals from one backend. A specialist publisher that has priced one commodity for decades is a focused provenance story: the whole apparatus exists to move this number.

Methodology note - every value is a spot price per troy ounce in the pair's currency, with daily points in the full-history series and one live point per pair at any moment. Timestamps arrive in the source's compact native encoding; request decoded dates in your sample if your pipeline should never see an encoded integer. Read the troy-ounce basis as invariant - gram and kilo views exist on request, not as separate columns to reconcile.

Completeness note - Datadory scores this record 6/10 against a catalog averaging 7.81 across 1,744 datasets, and the gap is structural, not accidental: three verified fields is the entire dictionary, chosen because the value here is longitudinal reach and currency breadth. Teams needing joined fundamentals per price point should pair this series with the survey and production datasets in the rail below rather than asking one source to be two things.