> This page is for Gateway. > For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.merge.dev/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.merge.dev/_mcp/server. # Create extraction POST https://api-gateway.merge.dev/v1/extractions Content-Type: application/json Read a piece of state and get one typed value per field you name. Reference: https://docs.merge.dev/merge-gateway/api-overview/extractions/create ## Authentication - `Authorization` header (bearer token, required) — Your production key sent as a bearer token. ## Request ### Body (application/json) This endpoint expects an ExtractionRequestSchema. - `model` (string, required) — Extraction model ID, for example `cloudflare/clairic`. - `fields` (ExtractionRequestSchemaFields, required) — Fields to extract, keyed by your names (up to 100). Each field's `type` sets the type of its value. - `state` (string, optional, default: ) — Text to read. Can be empty only when `images` are sent. - `images` (list of ExtractionRequestSchemaImagesItems, optional, nullable) — Up to 4 PNG, JPEG, or WebP images, as `data:` URLs or `{content_type, base64}` objects. Each up to 4 MiB decoded and 8 MiB in total. Remote URLs aren't accepted. Supported by `cloudflare/clairic` only. - `reference_date` (string, optional, nullable) — ISO date (`YYYY-MM-DD`) that anchors relative dates and dates written without a year. - `options` (ExtractionOptionsSchema, optional, nullable) — Extra output. Supported by `cloudflare/clairic` only. - `vendor` (string, optional, nullable) — Optional execution vendor pin. - `customer` (string, optional, nullable) — Optional customer UUID to scope the key, budget, and usage to. ## Response ### 200 Successful Response - `model` (string, required) — Model that answered. - `vendor` (string, required) — Vendor that served the request. - `answers` (map from string to any, required) — One typed answer per field, keyed by your field names. Each has `type` and `value` (null when the input doesn't determine it), plus per-type detail such as `confidence`, `span`, or a distribution. - `truncated` (boolean, required) — True when the input exceeded the model's context window and was cut from the end. - `usage` (ExtractionUsage, required) — Token usage for an extraction request. - `object` ("extraction", optional, default: extraction) - `state_text` (string, optional, nullable) — The text that `span` offsets index into, when `options.return_state_text` is true. ## Errors ### 422 Unprocessable Entity Error Validation Error - `detail` (list of ValidationError, optional) ## Types ### ExtractionRequestSchemaFields Fields to extract, keyed by your names (up to 100). Each field's `type` sets the type of its value. ### ExtractionRequestSchemaImagesItems - string - ExtractionImageSchema: ExtractionImageSchema ### ExtractionOptionsSchema Extra output for an extraction request. - `include_quantiles` (boolean, optional, nullable) — Add 99 percentiles to each `estimate` and `forecast` answer. - `return_state_text` (boolean, optional, nullable) — Return the text that `span` and `spans` offsets index into, as `state_text`. ### ExtractionUsage Token usage for an extraction request. - `input_tokens` (integer, required) - `output_tokens` (integer, required) - `total_tokens` (integer, required) - `passes` (integer, optional, nullable) — Passes the fields were split across. Each pass reads `state` again, so `input_tokens` counts it once per pass. - `cost` (double, optional, nullable) — Cost of the request in USD. ### ValidationError Single field-level validation error. `loc` is the JSON path to the offending field, `msg` is a human-readable explanation, and `type` is a stable machine-readable error code. - `loc` (list of ValidationErrorLocItems, required) - `msg` (string, required) - `type` (string, required) - `input` (any, optional) - `ctx` (ValidationErrorCtx, optional) ### ExtractionImageSchema An embedded image. - `content_type` (enum, required) - Allowed values: `image/png`, `image/jpeg`, `image/webp` - `base64` (string, required) — Base64 data, without a `data:` prefix. ### ValidationErrorLocItems - string - integer ### ValidationErrorCtx ## Examples **Request** ```json { "model": "string" } ``` **Response** ```json { "model": "string", "vendor": "string", "answers": {}, "truncated": true, "usage": { "input_tokens": 1, "output_tokens": 1, "total_tokens": 1, "passes": 1, "cost": 1.1 }, "object": "extraction", "state_text": "string" } ``` **SDK Code** ```python import requests url = "https://api-gateway.merge.dev/v1/extractions" payload = { "model": "string" } headers = { "Authorization": "Bearer ", "Content-Type": "application/json" } response = requests.post(url, json=payload, headers=headers) print(response.json()) ``` ```javascript const url = 'https://api-gateway.merge.dev/v1/extractions'; const options = { method: 'POST', headers: {Authorization: 'Bearer ', 'Content-Type': 'application/json'}, body: '{"model":"string"}' }; 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://api-gateway.merge.dev/v1/extractions" payload := strings.NewReader("{\n \"model\": \"string\"\n}") req, _ := http.NewRequest("POST", url, payload) req.Header.Add("Authorization", "Bearer ") 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://api-gateway.merge.dev/v1/extractions") http = Net::HTTP.new(url.host, url.port) http.use_ssl = true request = Net::HTTP::Post.new(url) request["Authorization"] = 'Bearer ' request["Content-Type"] = 'application/json' request.body = "{\n \"model\": \"string\"\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://api-gateway.merge.dev/v1/extractions") .header("Authorization", "Bearer ") .header("Content-Type", "application/json") .body("{\n \"model\": \"string\"\n}") .asString(); ``` ```php request('POST', 'https://api-gateway.merge.dev/v1/extractions', [ 'body' => '{ "model": "string" }', 'headers' => [ 'Authorization' => 'Bearer ', 'Content-Type' => 'application/json', ], ]); echo $response->getBody(); ``` ```csharp using RestSharp; var client = new RestClient("https://api-gateway.merge.dev/v1/extractions"); var request = new RestRequest(Method.POST); request.AddHeader("Authorization", "Bearer "); request.AddHeader("Content-Type", "application/json"); request.AddParameter("application/json", "{\n \"model\": \"string\"\n}", ParameterType.RequestBody); IRestResponse response = client.Execute(request); ``` ```swift import Foundation let headers = [ "Authorization": "Bearer ", "Content-Type": "application/json" ] let parameters = ["model": "string"] as [String : Any] let postData = JSONSerialization.data(withJSONObject: parameters, options: []) let request = NSMutableURLRequest(url: NSURL(string: "https://api-gateway.merge.dev/v1/extractions")! 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() ```