> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.reka.ai/vision/api-reference/search/search-videos/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.reka.ai/_mcp/server. # Search Videos POST https://vision-agent.api.reka.ai/v2/search Content-Type: application/json Reference: https://docs.reka.ai/vision/api-reference/search/search-videos ## Authentication - `X-Api-Key` header (required) — API key authentication via X-Api-Key header - `Authorization` header (bearer token, required) — Bearer token authentication via Authorization header ## Request ### Headers - `X-Enable-Overwrite-Metering` (string, optional, nullable) ### Body (application/json) This endpoint expects a V2SearchRequest. - `query` (string, required) — Search query — natural language for semantic, literal text for text - `strategy` (enum, optional) — 'semantic' (embedding similarity) or 'text' (literal substring match) - Allowed values: `semantic`, `text` - `fields` (list of enum, optional, nullable) — Which data to search. For text: 'transcript', 'captions' (default: both). Ignored for semantic. - Allowed values: `transcript`, `captions` - `video_ids` (list of string, optional, nullable) — Scope to specific video IDs - `group_ids` (list of string, optional, nullable) — Scope to videos in specific groups - `threshold` (enum, optional) — Confidence filter: high, medium, low, none. Only affects semantic strategy. - Allowed values: `high`, `medium`, `low`, `none` - `rerank` (boolean, optional, default: false) — Apply embedding-based reranking (semantic only, slightly slower) - `filter` (SearchFilter, optional, nullable) — Metadata filters (duration, width, height) - `datetime_from` (datetime, optional, nullable) — Start of an explicit UTC datetime range for semantic search - `datetime_to` (datetime, optional, nullable) — End of an explicit UTC datetime range for semantic search - `timestamp_from` (double, optional, nullable) — Start of an explicit in-video timestamp range in seconds - `timestamp_to` (double, optional, nullable) — End of an explicit in-video timestamp range in seconds - `page_limit` (integer, optional, default: 10) — Results per page (max 50) - `page_token` (string, optional, nullable) — Pagination token from a previous response - `include_report` (boolean, optional, default: false) — Generate LLM summary of results (semantic only) ## Response ### 200 Successful Response - `data` (list of V2SearchResult, required) - `search_pool` (SearchPool, required) - `next_page_token` (string, optional, nullable) - `report` (string, optional, nullable) — LLM-generated summary (when include_report=true, semantic only) ## Types ### SearchFilter Metadata filters applied to search results. - `id` (list of string, optional, nullable) — Filter to specific video IDs (same as top-level video_ids) - `duration` (NumberFilter, optional, nullable) — Filter by video duration in seconds - `width` (NumberFilter, optional, nullable) — Filter by video width in pixels - `height` (NumberFilter, optional, nullable) — Filter by video height in pixels ### V2SearchResult - `video_id` (string, required) — Video this moment is from - `score` (double, required) — Relevance score in [0, 1]. Text matches are binary; 1.0 is a sentinel. - `rank` (integer, required) — Position in results (1 = most relevant) - `video_name` (string, optional, nullable) — Video display name - `start` (double, optional, nullable) — Start time in seconds - `end` (double, optional, nullable) — End time in seconds - `caption` (string, optional, nullable) — AI description of the visual content - `transcript` (string, optional, nullable) — Spoken words in this segment - `thumbnail_url` (string, optional, nullable) — Preview frame URL (not yet available) - `video_url` (string, optional, nullable) — Presigned URL to the parent video ### SearchPool - `video_count` (integer, required) — Number of videos searched - `total_duration` (double, required) — Total duration searched (seconds) ### NumberFilter Numeric range filter with optional bounds. - `gte` (double, optional, nullable) - `lte` (double, optional, nullable) ## Examples **Request** ```json { "query": "person giving a presentation with charts" } ``` **Response** ```json { "data": [ { "video_id": "string", "score": 0.5, "rank": 1, "video_name": "string", "start": 1.1, "end": 1.1, "caption": "string", "transcript": "string", "thumbnail_url": "string", "video_url": "string" } ], "search_pool": { "video_count": 1, "total_duration": 1.1 }, "next_page_token": "string", "report": "string" } ``` **SDK Code** ```python import requests url = "https://vision-agent.api.reka.ai/v2/search" payload = { "query": "person giving a presentation with charts" } headers = { "X-Api-Key": "", "Content-Type": "application/json" } response = requests.post(url, json=payload, headers=headers) print(response.json()) ``` ```javascript const url = 'https://vision-agent.api.reka.ai/v2/search'; const options = { method: 'POST', headers: {'X-Api-Key': '', 'Content-Type': 'application/json'}, body: '{"query":"person giving a presentation with charts"}' }; try { const response = await fetch(url, options); const data = await response.json(); console.log(data); } catch (error) { console.error(error); } ``` ```go package main import ( "fmt" "strings" "net/http" "io" ) func main() { url := "https://vision-agent.api.reka.ai/v2/search" payload := strings.NewReader("{\n \"query\": \"person giving a presentation with charts\"\n}") req, _ := http.NewRequest("POST", url, payload) req.Header.Add("X-Api-Key", "") req.Header.Add("Content-Type", "application/json") res, _ := http.DefaultClient.Do(req) defer res.Body.Close() body, _ := io.ReadAll(res.Body) fmt.Println(res) fmt.Println(string(body)) } ``` ```ruby require 'uri' require 'net/http' url = URI("https://vision-agent.api.reka.ai/v2/search") http = Net::HTTP.new(url.host, url.port) http.use_ssl = true request = Net::HTTP::Post.new(url) request["X-Api-Key"] = '' request["Content-Type"] = 'application/json' request.body = "{\n \"query\": \"person giving a presentation with charts\"\n}" response = http.request(request) puts response.read_body ``` ```java import com.mashape.unirest.http.HttpResponse; import com.mashape.unirest.http.Unirest; HttpResponse response = Unirest.post("https://vision-agent.api.reka.ai/v2/search") .header("X-Api-Key", "") .header("Content-Type", "application/json") .body("{\n \"query\": \"person giving a presentation with charts\"\n}") .asString(); ``` ```php request('POST', 'https://vision-agent.api.reka.ai/v2/search', [ 'body' => '{ "query": "person giving a presentation with charts" }', 'headers' => [ 'Content-Type' => 'application/json', 'X-Api-Key' => '', ], ]); echo $response->getBody(); ``` ```csharp using RestSharp; var client = new RestClient("https://vision-agent.api.reka.ai/v2/search"); var request = new RestRequest(Method.POST); request.AddHeader("X-Api-Key", ""); request.AddHeader("Content-Type", "application/json"); request.AddParameter("application/json", "{\n \"query\": \"person giving a presentation with charts\"\n}", ParameterType.RequestBody); IRestResponse response = client.Execute(request); ``` ```swift import Foundation let headers = [ "X-Api-Key": "", "Content-Type": "application/json" ] let parameters = ["query": "person giving a presentation with charts"] as [String : Any] let postData = JSONSerialization.data(withJSONObject: parameters, options: []) let request = NSMutableURLRequest(url: NSURL(string: "https://vision-agent.api.reka.ai/v2/search")! as URL, cachePolicy: .useProtocolCachePolicy, timeoutInterval: 10.0) request.httpMethod = "POST" request.allHTTPHeaderFields = headers request.httpBody = postData as Data let session = URLSession.shared let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in if (error != nil) { print(error as Any) } else { let httpResponse = response as? HTTPURLResponse print(httpResponse) } }) dataTask.resume() ``` > One OpenAI-compatible API for Reka models and a curated selection of open models.