> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://respan.ai/docs/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://respan.ai/docs/_mcp/server. # Run span-02 POST https://api.respan.ai/api/v1/decisions Content-Type: application/json Answer typed questions about a state with span-02, Respan's decision model. Each question is `choice`, `noul` (yes/no), `score` or `number`, and each answer carries its probability distribution and confidence. One request can mix all four types. Set a thinking deadline with `reasoning.budget_ms`: span-02 thinks until then and answers with its best decision so far. The body is the System One format, so a System One client works by changing the URL. span-02 runs on Respan credits only and is priced per input and output token; thought tokens aren't billed. Streaming is unsupported. Reference: https://respan.ai/docs/apis/gateway/run-span-02 ## Authentication - `Authorization` header (bearer token, required) — Use your Respan API key for Respan API authentication. Enter only the Respan API key value; clients send Authorization: Bearer \. For /api/responses, provider credentials such as Perplexity, OpenAI, or Azure OpenAI go in Settings -> Providers or respan\_params.credential\_override in the request body, not in this authentication field. ## Request ### Body (application/json) This endpoint expects a Span2DecisionRequest. - `state` (Span2DecisionRequestState, required) — What the questions are about: text or any JSON value. - `questions` (map from string to Span2DecisionRequestQuestions, required) — Your question ids mapped to questions. Answers come back under the same ids. - `model` (string, optional, default: span-02) — Use `span-02` or `respan/span-02`. Omit for `span-02`. - `images` (list of string, optional) — Up to 4 images, as data URIs or base64. span-02 sees them before the text. - `reasoning` (Span2Reasoning, optional) — How span-02 thinks before it answers. The thought itself is never returned, and its tokens aren't billed. - `respan_params` (map from string to any, optional) — Respan parameters for the request log, such as `customer_identifier`. ## Response ### 200 Answers keyed by question id. The example is the captured response to the request shown; probabilities can vary slightly between calls. - `model` (enum, required) — The model that answered. - Allowed values: `span-02` - `answers` (map from string to Span2DecisionResponseAnswers, required) — Answers keyed by the ids in `questions`. - `usage` (Span2DecisionResponseUsage, optional) — Billed token counts. Thought tokens aren't billed. ## Errors ### 400 Bad Request Error Invalid request: an unknown model, `stream: true`, or questions the model rejects. Correct the request before retrying. - `map from string to any` ### 402 Payment Required Error The organization has no Respan credits. Add credits to call span-02. - `map from string to any` ### 403 Forbidden Error Missing, invalid or expired Respan API key. - `map from string to any` ### 429 Too Many Requests Error The organization's span-02 rate limit was reached. Retry with exponential backoff. - `map from string to any` ### 504 Gateway Timeout Error span-02 didn't answer in time. Retry; if it persists, send fewer questions or a shorter state. - `map from string to any` ## Types ### Span2DecisionRequestState What the questions are about: text or any JSON value. ### Span2DecisionRequestQuestions ### Span2Reasoning How span-02 thinks before it answers. The thought itself is never returned, and its tokens aren't billed. - `mode` (enum, optional, default: auto) — `auto` thinks only when span-02 is unsure, `on` thinks for every question, `off` never thinks. - Allowed values: `auto`, `on`, `off` - `budget_ms` (integer, optional, default: 0) — Thinking time for the request in milliseconds. span-02 thinks until then, then answers with its best decision so far. `0` means no deadline; a per-question cap on thought length still applies. ### Span2DecisionResponseAnswers ### Span2DecisionResponseUsage Billed token counts. Thought tokens aren't billed. - `input_tokens` (integer, optional) - `output_tokens` (integer, optional) ### TypeSafeChoiceQuestion Select an option from the supplied criteria. - `type` (enum, required) - Allowed values: `choice` - `criteria` (map from string to TypeSafeEntry, required) — Option names mapped to descriptions. Use null for an option without a description. - `instructions` (TypeSafeEntry, optional) — Question instructions or rubric descriptions. Accepts text, an object, an array, or null. See https://docs.typesafe.ai/primitives/advanced. ### TypeSafeNoulQuestion A yes/no judgment expressed as a probability. - `type` (enum, required) - Allowed values: `noul` - `instructions` (TypeSafeEntry, optional) — Question instructions or rubric descriptions. Accepts text, an object, an array, or null. See https://docs.typesafe.ai/primitives/advanced. - `criteria` (TypeSafeNoulQuestionCriteria, optional, nullable) — Optional guidance for yes and no. ### TypeSafeScoreQuestion Evaluate the state on an ordered rubric. - `type` (enum, required) - Allowed values: `score` - `criteria` (list of TypeSafeContent, required) — Rubric levels in order, starting at index 0. - `instructions` (TypeSafeEntry, optional) — Question instructions or rubric descriptions. Accepts text, an object, an array, or null. See https://docs.typesafe.ai/primitives/advanced. ### Span2NumberQuestion Ask for a number. span-02 answers with the number itself, so no options are listed. - `type` (enum, required) - Allowed values: `number` - `instructions` (TypeSafeEntry, required) — Question instructions or rubric descriptions. Accepts text, an object, an array, or null. See https://docs.typesafe.ai/primitives/advanced. ### TypeSafeChoiceAnswer - `type` (enum, required) - Allowed values: `choice` - `choice` (string, required) — Selected option name. - `probabilities` (map from string to double, required) - `confidence` (double, required) ### TypeSafeNoulAnswer - `type` (enum, required) - Allowed values: `noul` - `noul` (double, required) — Probability of yes: 0 means no, 1 means yes. ### TypeSafeScoreAnswer - `type` (enum, required) - Allowed values: `score` - `score` (double, required) — Probability-weighted level index; may be fractional. - `legend` (map from string to TypeSafeContent, required) — Level indices mapped to rubric descriptions. - `probabilities` (map from string to double, required) - `confidence` (double, required) ### Span2NumberAnswer - `type` (enum, required) - Allowed values: `number` - `number` (double, required, nullable) — The answer as a number, or null if the model's text is not a number. - `text` (string, required) — The answer as the model wrote it. - `confidence` (double, required) — How sure span-02 is of the answer. ### TypeSafeEntry Question instructions or rubric descriptions. Accepts text, an object, an array, or null. See https://docs.typesafe.ai/primitives/advanced. ### TypeSafeNoulQuestionCriteria Optional guidance for yes and no. - `true` (TypeSafeEntry, optional) — Question instructions or rubric descriptions. Accepts text, an object, an array, or null. See https://docs.typesafe.ai/primitives/advanced. - `false` (TypeSafeEntry, optional) — Question instructions or rubric descriptions. Accepts text, an object, an array, or null. See https://docs.typesafe.ai/primitives/advanced. ### TypeSafeContent Text or structured context supplied to Jev. ## Examples **Request** ```json { "state": { "customer_message": "My package arrived damaged and nobody answers my emails. I want a refund now.", "order": { "id": 123, "days_late": 4 } }, "questions": { "department": { "type": "choice", "instructions": "Which team should handle this ticket?", "criteria": { "returns": "Exchanges, wrong or damaged items", "shipping": "Delivery status, delays, lost packages", "billing": "Charges, invoices, payment problems" } }, "wants_refund": { "type": "noul", "instructions": "Does the customer ask for a refund?" }, "frustration": { "type": "score", "instructions": "How frustrated is the customer?", "criteria": [ "Calm or positive", "Mildly annoyed", "Angry or threatening to leave" ] }, "days_late": { "type": "number", "instructions": "How many days late is the order?" } }, "model": "span-02", "reasoning": { "mode": "auto" } } ``` **Response** ```json { "model": "span-02", "answers": { "department": { "type": "choice", "choice": "returns", "confidence": 0.969893, "probabilities": { "returns": 0.9799287, "shipping": 0.012929817, "billing": 0.00714149 } }, "wants_refund": { "type": "noul", "noul": 0.99997675 }, "frustration": { "type": "score", "score": 1.9921819, "legend": { "0": "Calm or positive", "1": "Mildly annoyed", "2": "Angry or threatening to leave" }, "probabilities": { "0": 0.00009271287, "1": 0.0076327343, "2": 0.9922746 }, "confidence": 0.9882728 }, "days_late": { "type": "number", "number": 4, "text": "4", "confidence": 0.9999089 } }, "usage": { "input_tokens": 580, "output_tokens": 1 } } ``` **SDK Code** ```python Respan Models_createDecisions_example import requests url = "https://api.respan.ai/api/v1/decisions" payload = { "state": { "customer_message": "My package arrived damaged and nobody answers my emails. I want a refund now.", "order": { "id": 123, "days_late": 4 } }, "questions": { "department": { "type": "choice", "instructions": "Which team should handle this ticket?", "criteria": { "returns": "Exchanges, wrong or damaged items", "shipping": "Delivery status, delays, lost packages", "billing": "Charges, invoices, payment problems" } }, "wants_refund": { "type": "noul", "instructions": "Does the customer ask for a refund?" }, "frustration": { "type": "score", "instructions": "How frustrated is the customer?", "criteria": ["Calm or positive", "Mildly annoyed", "Angry or threatening to leave"] }, "days_late": { "type": "number", "instructions": "How many days late is the order?" } }, "model": "span-02", "reasoning": { "mode": "auto" } } headers = { "Authorization": "Bearer ", "Content-Type": "application/json" } response = requests.post(url, json=payload, headers=headers) print(response.json()) ``` ```javascript Respan Models_createDecisions_example const url = 'https://api.respan.ai/api/v1/decisions'; const options = { method: 'POST', headers: {Authorization: 'Bearer ', 'Content-Type': 'application/json'}, body: '{"state":{"customer_message":"My package arrived damaged and nobody answers my emails. I want a refund now.","order":{"id":123,"days_late":4}},"questions":{"department":{"type":"choice","instructions":"Which team should handle this ticket?","criteria":{"returns":"Exchanges, wrong or damaged items","shipping":"Delivery status, delays, lost packages","billing":"Charges, invoices, payment problems"}},"wants_refund":{"type":"noul","instructions":"Does the customer ask for a refund?"},"frustration":{"type":"score","instructions":"How frustrated is the customer?","criteria":["Calm or positive","Mildly annoyed","Angry or threatening to leave"]},"days_late":{"type":"number","instructions":"How many days late is the order?"}},"model":"span-02","reasoning":{"mode":"auto"}}' }; try { const response = await fetch(url, options); const data = await response.json(); console.log(data); } catch (error) { console.error(error); } ``` ```go Respan Models_createDecisions_example package main import ( "fmt" "strings" "net/http" "io" ) func main() { url := "https://api.respan.ai/api/v1/decisions" payload := strings.NewReader("{\n \"state\": {\n \"customer_message\": \"My package arrived damaged and nobody answers my emails. I want a refund now.\",\n \"order\": {\n \"id\": 123,\n \"days_late\": 4\n }\n },\n \"questions\": {\n \"department\": {\n \"type\": \"choice\",\n \"instructions\": \"Which team should handle this ticket?\",\n \"criteria\": {\n \"returns\": \"Exchanges, wrong or damaged items\",\n \"shipping\": \"Delivery status, delays, lost packages\",\n \"billing\": \"Charges, invoices, payment problems\"\n }\n },\n \"wants_refund\": {\n \"type\": \"noul\",\n \"instructions\": \"Does the customer ask for a refund?\"\n },\n \"frustration\": {\n \"type\": \"score\",\n \"instructions\": \"How frustrated is the customer?\",\n \"criteria\": [\n \"Calm or positive\",\n \"Mildly annoyed\",\n \"Angry or threatening to leave\"\n ]\n },\n \"days_late\": {\n \"type\": \"number\",\n \"instructions\": \"How many days late is the order?\"\n }\n },\n \"model\": \"span-02\",\n \"reasoning\": {\n \"mode\": \"auto\"\n }\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 Respan Models_createDecisions_example require 'uri' require 'net/http' url = URI("https://api.respan.ai/api/v1/decisions") 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 \"state\": {\n \"customer_message\": \"My package arrived damaged and nobody answers my emails. I want a refund now.\",\n \"order\": {\n \"id\": 123,\n \"days_late\": 4\n }\n },\n \"questions\": {\n \"department\": {\n \"type\": \"choice\",\n \"instructions\": \"Which team should handle this ticket?\",\n \"criteria\": {\n \"returns\": \"Exchanges, wrong or damaged items\",\n \"shipping\": \"Delivery status, delays, lost packages\",\n \"billing\": \"Charges, invoices, payment problems\"\n }\n },\n \"wants_refund\": {\n \"type\": \"noul\",\n \"instructions\": \"Does the customer ask for a refund?\"\n },\n \"frustration\": {\n \"type\": \"score\",\n \"instructions\": \"How frustrated is the customer?\",\n \"criteria\": [\n \"Calm or positive\",\n \"Mildly annoyed\",\n \"Angry or threatening to leave\"\n ]\n },\n \"days_late\": {\n \"type\": \"number\",\n \"instructions\": \"How many days late is the order?\"\n }\n },\n \"model\": \"span-02\",\n \"reasoning\": {\n \"mode\": \"auto\"\n }\n}" response = http.request(request) puts response.read_body ``` ```java Respan Models_createDecisions_example import com.mashape.unirest.http.HttpResponse; import com.mashape.unirest.http.Unirest; HttpResponse response = Unirest.post("https://api.respan.ai/api/v1/decisions") .header("Authorization", "Bearer ") .header("Content-Type", "application/json") .body("{\n \"state\": {\n \"customer_message\": \"My package arrived damaged and nobody answers my emails. I want a refund now.\",\n \"order\": {\n \"id\": 123,\n \"days_late\": 4\n }\n },\n \"questions\": {\n \"department\": {\n \"type\": \"choice\",\n \"instructions\": \"Which team should handle this ticket?\",\n \"criteria\": {\n \"returns\": \"Exchanges, wrong or damaged items\",\n \"shipping\": \"Delivery status, delays, lost packages\",\n \"billing\": \"Charges, invoices, payment problems\"\n }\n },\n \"wants_refund\": {\n \"type\": \"noul\",\n \"instructions\": \"Does the customer ask for a refund?\"\n },\n \"frustration\": {\n \"type\": \"score\",\n \"instructions\": \"How frustrated is the customer?\",\n \"criteria\": [\n \"Calm or positive\",\n \"Mildly annoyed\",\n \"Angry or threatening to leave\"\n ]\n },\n \"days_late\": {\n \"type\": \"number\",\n \"instructions\": \"How many days late is the order?\"\n }\n },\n \"model\": \"span-02\",\n \"reasoning\": {\n \"mode\": \"auto\"\n }\n}") .asString(); ``` ```php Respan Models_createDecisions_example request('POST', 'https://api.respan.ai/api/v1/decisions', [ 'body' => '{ "state": { "customer_message": "My package arrived damaged and nobody answers my emails. I want a refund now.", "order": { "id": 123, "days_late": 4 } }, "questions": { "department": { "type": "choice", "instructions": "Which team should handle this ticket?", "criteria": { "returns": "Exchanges, wrong or damaged items", "shipping": "Delivery status, delays, lost packages", "billing": "Charges, invoices, payment problems" } }, "wants_refund": { "type": "noul", "instructions": "Does the customer ask for a refund?" }, "frustration": { "type": "score", "instructions": "How frustrated is the customer?", "criteria": [ "Calm or positive", "Mildly annoyed", "Angry or threatening to leave" ] }, "days_late": { "type": "number", "instructions": "How many days late is the order?" } }, "model": "span-02", "reasoning": { "mode": "auto" } }', 'headers' => [ 'Authorization' => 'Bearer ', 'Content-Type' => 'application/json', ], ]); echo $response->getBody(); ``` ```csharp Respan Models_createDecisions_example using RestSharp; var client = new RestClient("https://api.respan.ai/api/v1/decisions"); var request = new RestRequest(Method.POST); request.AddHeader("Authorization", "Bearer "); request.AddHeader("Content-Type", "application/json"); request.AddParameter("application/json", "{\n \"state\": {\n \"customer_message\": \"My package arrived damaged and nobody answers my emails. I want a refund now.\",\n \"order\": {\n \"id\": 123,\n \"days_late\": 4\n }\n },\n \"questions\": {\n \"department\": {\n \"type\": \"choice\",\n \"instructions\": \"Which team should handle this ticket?\",\n \"criteria\": {\n \"returns\": \"Exchanges, wrong or damaged items\",\n \"shipping\": \"Delivery status, delays, lost packages\",\n \"billing\": \"Charges, invoices, payment problems\"\n }\n },\n \"wants_refund\": {\n \"type\": \"noul\",\n \"instructions\": \"Does the customer ask for a refund?\"\n },\n \"frustration\": {\n \"type\": \"score\",\n \"instructions\": \"How frustrated is the customer?\",\n \"criteria\": [\n \"Calm or positive\",\n \"Mildly annoyed\",\n \"Angry or threatening to leave\"\n ]\n },\n \"days_late\": {\n \"type\": \"number\",\n \"instructions\": \"How many days late is the order?\"\n }\n },\n \"model\": \"span-02\",\n \"reasoning\": {\n \"mode\": \"auto\"\n }\n}", ParameterType.RequestBody); IRestResponse response = client.Execute(request); ``` ```swift Respan Models_createDecisions_example import Foundation let headers = [ "Authorization": "Bearer ", "Content-Type": "application/json" ] let parameters = [ "state": [ "customer_message": "My package arrived damaged and nobody answers my emails. I want a refund now.", "order": [ "id": 123, "days_late": 4 ] ], "questions": [ "department": [ "type": "choice", "instructions": "Which team should handle this ticket?", "criteria": [ "returns": "Exchanges, wrong or damaged items", "shipping": "Delivery status, delays, lost packages", "billing": "Charges, invoices, payment problems" ] ], "wants_refund": [ "type": "noul", "instructions": "Does the customer ask for a refund?" ], "frustration": [ "type": "score", "instructions": "How frustrated is the customer?", "criteria": ["Calm or positive", "Mildly annoyed", "Angry or threatening to leave"] ], "days_late": [ "type": "number", "instructions": "How many days late is the order?" ] ], "model": "span-02", "reasoning": ["mode": "auto"] ] as [String : Any] let postData = JSONSerialization.data(withJSONObject: parameters, options: []) let request = NSMutableURLRequest(url: NSURL(string: "https://api.respan.ai/api/v1/decisions")! 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() ```