MetalpriceAPI.com Metals Price API

Most metals feeds pick a lane: the London fixes, one exchange's floor, or a single white metal. MetalpriceAPI, operated by Lyker Labs LLC, publishes the whole counter instead. Its catalogue runs to roughly 300 symbols: the five precious metals - gold XAU, silver XAG, platinum XPT, palladium XPD, rhodium XRH - each carrying separate bid and ask variants; 15+ industrial metals from aluminium, cobalt and copper through iron ore, lithium, molybdenum, nickel, tin, uranium and zinc; India gold and silver rates broken out by city across 25+ cities; crypto and energy symbols; all of it expressible against 150+ currencies.

Two structural choices make the slice unusually pleasant to work with. First, quotes travel in both directions in the same payload: XAU-style keys give how much metal one unit of the base currency buys, while USDXAU-style reciprocals give the familiar currency-per-ounce reading - the documentation pairs XAU 0.00053853 with USDXAU 1857.0, and both figures describe the same instant. Second, the record families are typed rather than improvised: latest snapshots, dated daily history, hourly cuts, minute-level depth stamped from 24 April 2026, OHLC bars, arbitrary-amount conversions, window changes with start and end rates, and karat-keyed gold pricing from 24k down to 6k.

Within Datadory's catalog of 1,744 datasets across 159 viable industries, this record scores 8/10 for quality, with field definitions verified.

One documented snapshot flat, then the catalogue it resolves against:

# A documented rate snapshot, flat, exactly as the fields arrive

success   : true
timestamp : 1787312280              # 2026-08-21 11:38 UTC
base      : USD
rates     : {"XAU": 0.00053853, "USDXAU": 1857.0}
            # metal units per dollar  |  dollars per troy ounce - one payload, both directions

# The symbol catalogue the rates map resolves against
XAU  Gold       Troy Ounce       -BID / -ASK variants
XAG  Silver     Troy Ounce       -BID / -ASK variants
XPT  Platinum   Troy Ounce       -BID / -ASK variants
XPD  Palladium  Troy Ounce       -BID / -ASK variants
XRH  Rhodium    Troy Ounce       -BID / -ASK variants
XCU  Copper     standard ounce   industrial family
...  15+ industrial metals       lithium, molybdenum, neodymium, tellurium, uranium, zinc and more
...  India gold/silver rates     25+ cities, Ahmedabad through Ludhiana

Three things settle on sight. Shape: three envelope fields, one rates map, done - the whole record parses into a dataframe without a nesting fight. Direction: the XAU figure and the USDXAU figure are the same quote read both ways, so a chart labelled in dollars-per-ounce and a model expecting ounces-per-dollar draw from one row instead of maintaining a reciprocal by hand. Families: precious metals quote per troy ounce with bid and ask separated, industrial metals quote per standard ounce, and the India city symbols bring the counter price into the same map as the international ones - Ahmedabad and Ludhiana sitting beside XAU rather than in a separate country appendix.

Every other symbol in the catalogue repeats this anatomy, which is what makes cross-family boards cheap to assemble.

Nine documented fields, definitions verified during the August 2026 research pass. Three frame the envelope: success, timestamp (Unix epoch, second precision) and base, the currency everything else is expressed against. One carries the economics: the rates map, where XAU-style keys give metal units per unit of base currency and USDXAU-style keys give the reciprocal.

The rest arrive when the record type calls for them. Conversion records add query (the echoed amount with source and target), info (the quote rate applied and the moment it was struck) and result (the converted amount). Period-change records add start_rate, end_rate, change and change_pct, so a month-over-month or year-over-year read needs no differencing on your side. Karat records add the data block, gold priced from 24k down to 6k.

The dictionary below shows all of it. Nothing hides behind pagination, and the layout never changes symbol to symbol - which is why the same parser serves a gold ticker, a copper dashboard and an Ahmedabad counter-price board.

Deeper attributes ride the same envelope and fold into a delivery when the use case calls for them:

• Bid and ask variants - the -BID and -ASK forms of all five precious metals, separating the two sides of the quote so spreads are measured rather than inferred from a single mid.
• OHLC bar records - open, high, low and close aggregated per interval, giving session-level structure that raw snapshots flatten away.
• Minute-level depth - records stamped from 24 April 2026 onward, the right grain for microstructure and event-window work.
• The full symbol catalogue - roughly 300 entries with names and units, filterable by family, so scoping happens once instead of per request.
• Crypto and energy symbols - sitting inside the same rates map as the metals, useful when a dashboard needs one feed for the whole commodity strip.

They ship mapped to your scope once the sample names the symbols and windows that matter.

Geography - one global quoting universe rather than venue-by-venue fragmentation, expressed against 150+ currencies. The India layer is the standout: gold and silver broken out by city across 25+ Indian cities, from Ahmedabad through Ludhiana, which turns local-market pricing from a shop-by-shop survey into a lookup. Crypto and energy symbols widen the frame past metals entirely.

Temporal - latest snapshots sit beside dated daily records, hourly cuts and minute-level depth stamped from 24 April 2026. Historical windows are delivered up to 365 days at a time and chain cleanly for longer arcs, so a decade of daily gold history is consecutive deliveries rather than a different product.

Granularity - per-symbol rate snapshots, with units declared by family: troy ounce for the precious metals, standard ounce for the industrial side. Bid and ask travel separately for the five precious metals. Above the snapshots sit OHLC bars and amount-level conversions, so aggregation and conversion are record types rather than homework.

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

Records arrive flattened: the envelope resolved, the rates map exploded one row per symbol with both quote directions typed, conversion and change records keyed to the window you name. Name the symbols, the currency and the resolution when you ask for the sample - a gold-and-silver hourly cut, the full precious five with bid and ask, or the industrial families in your reporting currency. The sample ships first either way; the ongoing feed lands on whatever cadence your workflow runs, shaped to the scope you named rather than as an undifferentiated dump.

A breadth-first metals catalogue earns its keep in five specific jobs:

• Jewellery and retail pricing - karat ladders from 24k down to 6k plus the India city layer turn headline spot into counter prices without guessing at a local markup.
• FX-aware metals analysis - the same metal against 150+ currencies answers whether a move belongs to the metal or to the dollar, in one lookup instead of two sources.
• Multi-metal dashboards - five precious metals and 15+ industrial ones sharing a schema means a board spanning gold, copper, lithium and zinc assembles from a single feed rather than a stitch job.
• Backtesting and feature engineering - dated dailies, hourly cuts and minute-level depth give strategies and models regularly shaped, uniformly typed inputs at whatever resolution the thesis needs.
• Spread and liquidity work - separate bid and ask variants for all five precious metals make half-spreads a measurement instead of an assumption.

Start from the precious metals & minerals data hub, then see how this record pairs against official statistics in Indian Bureau of Mines statistics vs MetalpriceAPI.

Investors & Quant Researchers get two-sided quotes and multi-resolution history across the precious five, with reciprocal quoting handled upstream - see investors quants use cases. Developers & Data-Product Builders get the smallest plausible integration surface: a nine-field envelope that parses once and powers tickers, alerts and converters regardless of symbol - see developers builders use cases. Data Scientists & ML Engineers get typed panels across 20+ metal families with no unit reconciliation step - see data scientists use cases. Market Researchers & Consultants get metals context in whatever currency the client reports in - see market researchers use cases. Competitive Intelligence & Product Teams get stable pricing baselines for bullion-adjacent offerings - see competitive intel use cases. E-commerce Operators get live mark-to-market for gold and silver SKUs, with karat ladders matching how jewellery buyers shop - see e-commerce operators use cases.

Provenance note - published by MetalpriceAPI, operated by Lyker Labs LLC, whose catalogue spans precious and industrial metals, India city rates, crypto, energy and 150+ currencies in one symbol namespace. One publisher across the whole strip is why the envelope holds identically from XAU to ZNC. The Kitco source profile covers another widely quoted metals reference in the vertical.

Methodology note - mind the two unit systems: precious metals quote per troy ounce, industrial metals per standard ounce, and the difference matters the moment a board mixes families. Quotes arrive in both directions per symbol, so decide once whether your models consume currency-per-unit or units-per-currency and stay consistent - mixing directions mid-series is the classic way a metals panel grows a phantom trend. Where a benchmark audit matters rather than an indicative read, pair this record with the auction fixes in the cards below.

Completeness note - Datadory scores this record 8/10 against a catalog mean of 7.81 across 1,744 datasets, above the middle of the 27-record precious metals & minerals slice. Breadth is the asset here: few sources put rhodium, lithium, uranium and an Ahmedabad gold rate in one namespace. Depth on fundamentals lives in the destination feeds joined through once a symbol list is settled - start from the ranked best precious metals & minerals datasets shortlist, then pull a sample.