AuthorizationstringrequiredheaderRequired Bearer API key. Programmatic access to the World API needs a key; browsing nosible.world in a browser does not.
Example: Bearer nos_sk_...
Find geographic locations mentioned or supported by the entities in one text.
Available only in V2.0-preview. Send one text and receive the same schema, model, text, and predictions envelope as ticker tagging and the other Model APIs. Each prediction contains a resolved location, from direct place names or supported Atlas graph associations. No matches produces an empty predictions array. Coordinates describe a representative point for each resolved place; they are not precise event coordinates. Location predictions contain no scores or evidence chains.
These are the fields you can build against. Nested names use dot notation; optional sections are called out in their descriptions.
schemastringStable response schema shared with the other text model endpoints.
Example: nosible_model_classification_v1
modelstringPinned geography tagger engine revision.
Example: NOSIBLE/geo-tagger-v2.0-preview@revision:c9fe46347e5422548c0e0780ddbc6252284619a1
textstringThe exact input text, preserved without trimming.
Example: A concert took place at Newlands Stadium in Cape Town, South Africa.
predictionsobject[]Resolved geographic locations. An empty array is a successful no-match result.
Example: [{ continent, region, country, lat, lng, source, matched_on, entity_type }]
predictions[].continentstring | nullContinent containing the resolved place, when known.
Example: Africa
predictions[].regionstring | nullState or province associated with the resolved place, when known.
Example: null
predictions[].countrystring | nullCountry containing the resolved place, when known.
Example: South Africa
predictions[].latnumberRepresentative latitude in decimal degrees, from −90 to 90.
Example: -29
predictions[].lngnumberRepresentative longitude in decimal degrees, from −180 to 180.
Example: 24
predictions[].sourcestringLocation source. Text matching and supported graph associations use ner; the World location schema also supports llm.
Example: ner
predictions[].matched_onstringThe text or supporting entity matched to this location.
Example: South Africa
predictions[].entity_typestring (optional)Type of the supporting named entity, when available.
Example: GPE
invalid_json — Request body is not valid UTF-8 JSON.api_key_required / invalid_api_key — A valid NOSIBLE bearer API key is required.request_timeout — Request body did not finish within 5 seconds.request_too_large — JSON request body exceeds the endpoint's byte limit.validation_error — Text is missing, blank, too long, or contains unsupported request fields; a batch must also satisfy the item and total-text limits.rate_limited / model_rate_limited — Caller or upstream rate limit exceeded. Respect Retry-After when present.model_invalid_response — The geocoder returned an invalid location or an incomplete batch. This is distinct from a successful empty predictions array.model_upstream_error / model_upstream_unreachable — The geocoder rejected the request or could not be reached.model_not_configured — The geocoder is not configured for this deployment.api_key_auth_not_configured / api_key_auth_unavailable / api_key_metering_unavailable — Authentication or quota metering is temporarily unavailable.model_upstream_timeout — The geocoder did not respond within 10 seconds. No automatic retries are made.not_found — Ticker and Geo tagging are unavailable in V1.2. Use the /api/v2.0-preview/models/ routes.