market_cap_usd_gte / _lteUSDConverted with a weekly FX snapshot; no rate means null, and those companies sit out of size filters entirely rather than being converted wrong

momentum5 params

ret_1m_gte / _ltepercent10.55 means +10.55%

ret_3m_gte / _ltepercentSame unit family as the other returns

ret_6m_gte / _ltepercent—

ret_12m_gte / _ltepercentHistory starts 2025-07-14, so twelve months is the longest window that exists

include_suspect_returnsbooleanDefault false. Opens the gate to rows where price return and market-cap change disagree — usually an unadjusted split

valuation5 params

pe_ttm_gte / _lteratioNull when lossmaking, so loss-makers cannot be filtered in or out here

pb_gte / _lteratioNull when equity is negative

ps_ttm_gte / _lteratio—

fcf_yield_gte / _ltefraction0.05 = 5%, can be negative

dividend_yield_gte / _ltefraction0.03 = 3%

profit & leverage8 params

gross_margin_gte / _ltefraction0.4 = 40%. Null by design for banks, insurers and REITs

operating_margin_gte / _ltefraction0.4 = 40%

net_margin_gte / _ltefraction0.4 = 40%

return_on_equity_gte / _ltefraction—

debt_to_equity_gte / _lteratio—

revenue_growth_yoy_gte / _ltefractionYear over year, as a fraction

eps_growth_yoy_gte / _ltefractionYear over year, as a fraction

is_lossmakingbooleanThe filter a null PE cannot express

guidance6 params

guidance_raises_90d_gteintegerCounts the number going UP — a bigger cost or tax rate is bad news

guidance_cuts_90d_gteintegerCounts the number going down

guidance_better_90d_gteintegerCounts direction of the news. This is the one you usually want

guidance_worse_90d_gteintegerCounts direction of the news

guidance_better_sinceYYYY-MM-DDLast good print after this date

guidance_worse_sinceYYYY-MM-DDLast bad print after this date

events · 90d8 params

event_familycsv, any matchEach regulator's own vocabulary; the words are never mapped across markets

event_sourceenumsec_8k, japan_tdnet, korea_dart, hkex, news. korea_dart is the only door into Korea

news_event_typecsv, any matchModel-read labels — a separate vocabulary from event_family

rollup_familycsv, any matchThe only cross-market axis: 34 families where SEC earnings_release meets TDnet earnings_flash

event_family_sinceYYYY-MM-DDEach label carries its own clock

news_event_type_sinceYYYY-MM-DDEach label carries its own clock

rollup_family_sinceYYYY-MM-DDEach label carries its own clock

event_sinceYYYY-MM-DDLatest event from ANY source — it cannot narrow a single label

holdings & governance5 params

short_percent_float_gte / _lte0–100US only

days_to_cover_gtedaysUS only

insider_net_90dbuying / sellingOpen-market buys and sales only — gifts and grants are excluded on purpose

insider_ownership_pct_gte / _lte0–100The proxy's "all directors and officers as a group" line

ceo_total_comp_gte / _lteUSDCEO total compensation from the latest proxy

taiwan2 params

tw_foreign_held_pct_gte / _lte0–100Foreign ownership share, daily since 2024-01-02

tw_revenue_yoy_pct_gte / _ltepercentLatest monthly revenue, year over year

product & relations5 params

find_termsstring · freeSearch the 60,620-term product vocabulary before you filter on it

makes_termcsv"It makes this"

related_to_termcsv"It is on this value chain" — direction mostly unknown, read as related, not supplier

peer_ofcompany_id / tickerSelf-reported compensation peer group, 29,308 edges, matched both ways

competitor_ofcompany_id / tickerNamed in 10-K Item 1, 7,980 edges. An unmatched identifier errors instead of returning empty

paging & sorting5 params

sortone of 27 keysDefault market_cap_usd

orderasc / desc—

limitinteger ≤ 1000Constrains the whole response, not each period

cursoropaqueKeyset pagination, echoed back as next_cursor

list_facetsboolean · freeReturns filterable values and counts instead of companies

One screen row · NVDA

{

"company_id": "cmp_006294", "company_name": "NVIDIA Corporation",

"country": "US", "sector": "Technology", "industry": "Semiconductors",

"primary_ticker": "NVDA", "primary_exchange": "NASDAQ",

"cik": "0001045810", "has_cik": true, "n_listings": 2,

"market_cap_local": "5533287450000", "market_cap_currency": "USD",str

"market_cap_usd": 5533287450000, "market_cap_as_of": "2026-09-03",

"market_cap_source": "daily_market_caps",

"fx_as_of": "2026-08-24", "price_as_of": "2026-09-03",

"ret_1m": 10.554587688734, "ret_3m": 6.50349650349651,percent

"ret_6m": 27.0295818505338, "ret_12m": 34.0669014084507,percent

"return_suspect": false,

"short_percent_float": "1.23", "short_percent_outstanding": "1.18",0–100 · str

"days_to_cover": "2.52", "short_interest_change_pct": "-2.29",str

"short_interest_as_of": "2026-08-14",

"guidance_raises_90d": 0, "guidance_cuts_90d": 0,

"guidance_better_90d": 0, "guidance_worse_90d": 0,

"guidance_last_at": null,

"event_sources_90d": ["news", "sec_8k"],

"event_families_90d": ["debt_obligation", "earnings_release",

"material_agreement", "officer_director_change", "shareholder_vote"],

"news_event_types_90d": ["analyst", "capital", "capital_return",

"competition", "contract", "earnings", "expansion", "filing",

"guidance", "legal", "macro", "management", "merger_acquisition",

"narrative_change", "operating_metric", "other", "ownership",

"partnership", "pricing", "product", "regulatory",

"share_price_move", "strategy"],

"rollup_families_90d": ["analyst", "capital", "capital_return",

"competition", "contract", "corporate_action", "earnings",

"expansion", "..."],

"event_filings_90d": 5, "news_events_90d": 986,

"event_last_at": "2026-09-04T19:50:08Z",

"filing_last_at": "2026-08-26T20:21:19Z",

"news_event_last_at": "2026-09-04T19:50:08Z",

"product_terms": 98, "related_terms": 60,

"peer_companies": 18, "competitor_companies": 19,

"ceo_name": "Jen-Hsun Huang", "ceo_total_comp": "36343830",str

"ceo_comp_fiscal_year": 2026, "insider_ownership_pct": "3.94",0–100 · str

"tw_foreign_held_pct": null, "tw_revenue_yoy_pct": null,0–100

"fundamentals_period": "2026-04-26", "fundamentals_grain": "ttm",

"fundamentals_source": "sec_us_gaap", "fundamentals_currency": "USD",

"fundamentals_filed_at": "2026-05-20",

"fundamentals_snapshot_date": "2026-08-17",

"revenue": 253491000000, "net_income": 159613000000,

"operating_income": 162285000000, "shareholders_equity": 195474000000,

"total_assets": 259474000000, "total_debt": 8470000000,

"cash_and_equivalents": 13237000000, "free_cash_flow": 119076000000,

"eps_diluted": 6.53,

"gross_margin": 0.741454331711974, "operating_margin": 0.640200243795638,fraction

"net_margin": 0.629659435640713,fraction

"return_on_equity": 0.904920244807479,fraction

"return_on_assets": 0.684627378146467,fraction

"return_on_invested_capital": 0.790667648377858,fraction

"debt_to_equity": 0.0433305708176023,fraction

"revenue_growth_yoy": 0.706837693162307,fraction

"earnings_growth_yoy": 1.07899809831453,fraction

"eps_growth_yoy": 1.08917197452229,fraction

"free_cash_flow_growth_yoy": 0.652364564831261,fraction

"book_value_per_share": 8.0338659131775,

"free_cash_flow_per_share": 4.89395324942204,

"payout_ratio": null,fraction

"is_lossmaking": false, "equity_negative": false,

"fundamentals_stale": false,

"pe_ttm": 34.6668971199088, "pb": 28.307025230977,

"ps_ttm": 21.8283388759364,

"fcf_yield": 0.0215199374830961, "dividend_yield": nullfraction

}

02 — Research deeper without wondering what your agent missed

Give your AI the financial history before asking it to reason

Complete historyLess retrievalDeeper analysis

Prompt

Walk me through NVIDIA’s revenue and margin progression over the last eight quarters. Which segments drove the acceleration?

Under the hoodhistoricals · 10-K / 10-Q

income_statements33 columns

Profit and loss as filed, straight from the filing’s XBRL.

GET/v1/financials/income-statements?ticker=NVDA&period=quarterly

keys & metadataPK cik + report_period + period

cikKey. SEC number

report_periodKey. Period end

periodKey. annual, quarterly or ttm — required, no default

fiscal_periodQ1…Q4 or FY

calendar_dateAligned to the natural quarter end; often null for off-cycle fiscal years

tickerIssuer symbol

currencyFiling currency — never converted to USD

accounting_standardus_gaap or ifrs, tagged per row

accession_numberBack to filings

filing_urlSource document directory

revenue & cost

revenueNet revenue

cost_of_revenueCost of revenue

gross_profitGross profit

operating_expenseTotal operating expense

research_and_developmentR&D

selling_general_and_administrative_expensesSG&A

operating_incomeOperating income

ebitEarnings before interest and tax

below the line

interest_expenseInterest expense

nonoperating_income_expenseNon-operating income and expense

income_tax_expenseIncome tax

net_incomeNet income

consolidated_incomeConsolidated income including minorities

net_income_common_stockAttributable to common stock

net_income_continuing_operationsNull means undisclosed, not zero

net_income_discontinued_operationsA negative bottom line can be a disposal, not the business

net_income_non_controlling_interestsMinority interests

per share & shares

earnings_per_shareBasic EPS

earnings_per_share_dilutedDiluted EPS

weighted_average_sharesWeighted average shares

weighted_average_shares_dilutedDiluted weighted average shares

dividends_per_common_shareDividend per share

preferred_dividends_impactPreferred dividend impact

NVDA FY2027 Q1 · quarterlyaccession 0001045810-26-000052

{
  "cik": 1045810, "ticker": "NVDA", "period": "quarterly",
  "report_period": "2026-04-26", "fiscal_period": "Q1",
  "calendar_date": null,
  "currency": "USD", "accounting_standard": "us_gaap",

  "revenue":         81615000000,
  "cost_of_revenue": 20458000000,
  "gross_profit":    61157000000,
  "operating_expense": 7621000000,
  "research_and_development": 6321000000,
  "selling_general_and_administrative_expenses": 1300000000,
  "operating_income": 53536000000,
  "ebit":             53536000000,

  "interest_expense": 102000000,
  "nonoperating_income_expense": 16367000000,
  "income_tax_expense": 11582000000,
  "net_income":          58321000000,
  "consolidated_income": 58321000000,
  "net_income_common_stock": 58321000000,
  "net_income_discontinued_operations": 0,
  "net_income_non_controlling_interests": null,

  "earnings_per_share": 2.4,
  "earnings_per_share_diluted": 2.39,
  "weighted_average_shares":         24286000000,
  "weighted_average_shares_diluted": 24391000000,
  "dividends_per_common_share": 0.01,
  "preferred_dividends_impact": 0,

  "accession_number": "0001045810-26-000052",
  "filing_url": "https://www.sec.gov/Archives/edgar/data/
                 1045810/000104581026000052"
}

With detailed reporting segments

See revenue and profit split by business, product line and region — where the growth actually came from

income_statement → revenue → segment

statement3 blocks

line item3 filed

dimension4 possible

{ label, value }leaves

03 — Know your agent didn’t silently miss something

Missing data should never become a plausible answer

Correct definitionsExplicit gapsVerifiable answers

Prompt

Compare R&D intensity across these companies over five years. Flag periods where the underlying data isn’t comparable or wasn’t disclosed.

What travels with every numberresearch_and_development_expense

value3,374,000,000USD, as reported — never scaled silently

fiscal_periodQ2 FY2026Company fiscal calendar, not calendar quarter

accounting_standardUS-GAAPIFRS filers flagged before any cross-company comparison

source10-Q · 0001045810-26-000041Accession number links straight to the filing

filed_at2026-03-25Point-in-time: what was knowable, and when

peer_valuenull · not disclosedMissing stays missing — never coerced to 0

04 — Understand who is behind the stock

See how ownership is changing — without mixing incompatible signals

InstitutionsInsidersMajor holdersShort interest

Prompt

Who has been building a position in this company? Are insiders buying or selling, and how has institutional ownership changed?

Under the hoodownership · 13F · Form 4

ownership12 columns · ~45,896 rows

Beneficial ownership from DEF 14A. Roughly 2,770 US companies by market cap, refreshed annually with the proxy — the latest snapshot only, with no history.

GET/v1/companies/cmp_006294/proxy-ownership

columnsPK id

idPrimary key

company_idBack to the registry

tickerIssuer symbol

owner_nameHolder

owner_roleRole, e.g. President and CEO

owner_kindinsider, insider_group (the directors-and-officers total) or major_holder (5%+)

is_groupWhether this row is a total

security_classShare class — dual-class companies split into rows

shares_ownedShares held

percentShare of class, 0–100

percent_sourcereported (printed by the company) or computed (divided by us)

confidenceExtraction confidence: high, medium or low

NVDA · first three rowsreported vs computed

[
  { "id": "130300", "company_id": "cmp_006294", "ticker": "NVDA",
    "owner_name": "Jen-Hsun Huang",
    "owner_role": "President and CEO",
    "owner_kind": "insider", "is_group": false,
    "security_class": "Common Stock",
    "shares_owned": "870604104", "percent": "3.58",
    "percent_source": "reported", "confidence": "high" },

  { "id": "130301", "owner_name": "Colette M. Kress",
    "owner_role": "EVP and CFO",
    "owner_kind": "insider",
    "shares_owned": "4443523", "percent": "0.018",
    "percent_source": "computed", "confidence": "medium" },

  { "id": "130302", "owner_name": "Ajay K. Puri",
    "owner_role": "EVP, Worldwide Field Operations",
    "owner_kind": "insider",
    "shares_owned": "3266897", "percent": "0.013",
    "percent_source": "computed", "confidence": "medium" }
]

Built for agents that need the facts, not another webpage

Every field carries the context required to interpret it correctly — source, date, unit, definition and provenance where applicable. Your agent shouldn’t spend its context searching the internet and guessing which number is right.

Give it the data. Spend the tokens on reasoning.

Start for free

Fields and row counts measured 2026-09-04 against the primary database and live endpoint payloads. Coverage numbers move as backfills land — quote the coverage block of your own response.

Docs ETF data Live events