Search a collection by meaning
Embeds the query text and returns the nearest documents by `vector_field`. Works in any language the embedding model understands; combine with `filter_by` for facets.
Операция для поискового ключа
Embeds the query text and returns the nearest documents by vector_field. Works in any language the embedding model understands; combine with filter_by for facets.
Authorization
api_key_header Every request carries an organization API key. A key has one of two scopes. A full key reaches every operation in this document and belongs on your server. A search key is the one to ship in a browser widget or a storefront page: it reaches only the operations marked x-nova-key-scope: search — searching, federated search, recommendations, event ingestion, and the visitor-facing AI operations (semantic and image search, voice transcription and voice search, assistant chat, and the visitor's own conversation). It cannot write documents or read or change configuration; any other operation answers 403. AI operations are paid in credits from the organization's wallet whichever key calls them (402 when the balance runs out), and a search key needs X-NOVA-SESSION-ID on them: the session is what a conversation belongs to, and what the per-session allowance (429) is counted against. A key of either scope may additionally carry restrictions set in the console: an allowlist of collections (403 elsewhere), a forced filter_by per collection that is ANDed into every search (document-level access control), an allowlist of fields (include_fields is limited to it and a request naming any other field in query_by, facet_by, sort_by, filter_by or highlight_fields answers 400), an IP/CIDR allowlist and an Origin/Referer allowlist (403), and a per-key rate limit (429 with Retry-After).
In: header
Path Parameters
The name of the collection
Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
Response Body
application/json
application/json
application/json
application/json
application/json
application/json
application/json
curl -X POST "https://example.com/collections/string/documents/search/semantic" \ -H "Content-Type: application/json" \ -d '{ "vector_field": "string", "q": "string" }'{ "_trace": {}, "facet_counts": [ { "counts": [ { "count": 0, "highlighted": "string", "value": "string", "parent": {} } ], "field_name": "string", "sampled": true, "stats": { "max": 0.1, "min": 0.1, "sum": 0.1, "total_values": 0, "avg": 0.1 } } ], "found": 0, "found_docs": 0, "search_time_ms": 0, "out_of": 0, "search_cutoff": true, "page": 0, "grouped_hits": [ { "found": 0, "group_key": [ null ], "hits": [ { "highlights": [ { "field": "company_name", "snippet": "<mark>Stark</mark> Industries" } ], "document": { "id": "124", "company_name": "Stark Industries", "num_employees": 5215, "country": "USA" }, "text_match": 1234556 } ] } ], "hits": [ { "highlights": [ { "field": "company_name", "snippet": "<mark>Stark</mark> Industries" } ], "document": { "id": "124", "company_name": "Stark Industries", "num_employees": 5215, "country": "USA" }, "text_match": 1234556 } ], "request_params": { "collection_name": "string", "first_q": "string", "q": "string", "per_page": 0, "voice_query": { "transcribed_query": "string" } }, "conversation": { "answer": "string", "conversation_history": [ {} ], "conversation_id": "string", "query": "string" }, "union_request_params": [ { "collection_name": "string", "first_q": "string", "q": "string", "per_page": 0, "voice_query": { "transcribed_query": "string" } } ], "metadata": {}, "parsed_nl_query": { "parse_time_ms": 0, "generated_params": {}, "augmented_params": {}, "llm_response": {} }, "query_rewrite": { "strategy": "layout", "original_q": "string", "q": "string" }, "rerank": { "applied": true, "model": "bge-reranker-base", "scored": 0, "ms": 0, "skipped": "string" }, "redirect": { "url": "string", "query": "string" }, "suggestions": [ { "query": "string", "source": "manual", "filter_by": "string", "facet": { "field": "string", "value": "string", "count": 0 } } ], "query_understanding": { "intent": "navigational", "relation": "SUBSTITUTE", "strategy": "string", "policy": "string", "search_text": "string", "protected_tokens": [ "string" ], "entities": [ { "entity_id": "string", "type": "brand", "canonical": "string", "alias": "string", "confidence": 0 } ], "constraints": [ { "kind": "string", "op": "eq", "value": null, "unit": "string", "currency": "string", "hardness": "hard", "text": "string", "filter_by": "string" } ], "confidence": 0 }, "correction": { "state": "NO_CORRECTION", "applied": { "q": "string", "type": "RAW", "score": 0, "found": 0 }, "suggestions": [ { "q": "string", "type": "RAW", "score": 0, "found": 0 } ], "reason": "string" }, "refinements": [ { "id": "string", "type": "facet", "label": "string", "query": "string", "filter_by": "string", "field": "string", "value": "string", "count": 0, "score": 0, "dimension": "string" } ]}Embed text POST
Turns text into 1024-dimensional multilingual vectors (`@cf/baai/bge-m3`). Use it to fill a `float[]` field of your documents so `searchSemantic` and `searchByImage` can match them; the same model embeds queries and image descriptions.
Search a collection by image POST
Finds documents whose `vector_field` is closest to the given image — an uploaded `media_id`, an `image_url`, or inline `image_base64`. The image is embedded the same way uploads are, so results are consistent with `uploadMedia`.