> ## Documentation Index > Fetch the complete documentation index at: https://docs.opper.ai/llms.txt > Use this file to discover all available pages before exploring further. # Decisions > OpenAI Decisions-compatible synchronous evaluation. Send `model`, `input` (a string, or user messages with `input_text` and inline base64 `input_image` parts) and typed `questions` (`predicate`, `choice`, `score`); `answers` come back in question order with probabilities and confidence; a question the model declines comes back as a `refusal` answer. A bare model name is OpenAI's, e.g. `gpt-6-luna`, which is called natively and answers with OpenAI's body unchanged. Other decision models are addressed by their full catalog ID, e.g. `typesafe/jev-latest`, `opper/kev-4b` or `greenpt/metis` (list available endpoints with `GET /v3/models?type=evaluation`); those are reached by translation to System One and accept text input only. GreenPT Metis receives all 1–32 questions together in one upstream request. Translation preserves question order, probabilities and confidence, but joins text-message boundaries and does not expose native impact, score legends or additional answer metadata; use the System One endpoint for the native contract. Billed on input tokens. Uses a project-scoped Opper API key, Comply policies and provider entitlements. Cost and generation ID are returned in headers. Streaming is unsupported. See the [classification and scoring guide](/build/gateway/evaluations) for working examples, model discovery, and usage and billing behavior. ## OpenAPI ````yaml post /v3/compat/decisions openapi: 3.1.0 info: description: Schema-driven generative API that orchestrates LLM-powered workflows. title: Task API version: 3.0.0 servers: - description: Production url: https://api.opper.ai - description: Local development url: http://localhost:8080 security: - BearerAuth: [] tags: - description: Schema-driven function management and execution name: Functions - description: OpenAI-compatible chat completions name: Chat - description: OpenAI Responses API compatible endpoint name: Responses - description: Google-compatible interactions endpoint name: Interactions - description: Model registry and capabilities name: Models - description: Synchronous image generation name: Images - description: Text-to-speech and speech-to-text name: Audio - description: Asynchronous video generation name: Videos - description: Reusable file storage for media inputs and generated outputs name: Files - description: Async generation status and downloads name: Artifacts - description: OpenAI-compatible embeddings name: Embeddings - description: Recorded HTTP request/response generations name: Generations - description: System health and status name: System - description: Roundtable endpoint — fan out a query to multiple LLMs and combine results name: Roundtable - description: Web search, fetch, and other utility tools name: Tools - description: Caller identity, credits, and usage name: Account - description: >- Programmatic project and API-key management. Authenticates with an `op-mak-…` management token; available on the control_plane and enterprise plans. name: Management paths: /v3/compat/decisions: post: tags: - Compatibility summary: Decisions API description: >- OpenAI Decisions-compatible synchronous evaluation. Send `model`, `input` (a string, or user messages with `input_text` and inline base64 `input_image` parts) and typed `questions` (`predicate`, `choice`, `score`); `answers` come back in question order with probabilities and confidence; a question the model declines comes back as a `refusal` answer. A bare model name is OpenAI's, e.g. `gpt-6-luna`, which is called natively and answers with OpenAI's body unchanged. Other decision models are addressed by their full catalog ID, e.g. `typesafe/jev-latest`, `opper/kev-4b` or `greenpt/metis` (list available endpoints with `GET /v3/models?type=evaluation`); those are reached by translation to System One and accept text input only. GreenPT Metis receives all 1–32 questions together in one upstream request. Translation preserves question order, probabilities and confidence, but joins text-message boundaries and does not expose native impact, score legends or additional answer metadata; use the System One endpoint for the native contract. Billed on input tokens. Uses a project-scoped Opper API key, Comply policies and provider entitlements. Cost and generation ID are returned in headers. Streaming is unsupported. operationId: createDecision parameters: - description: >- Function name for tracing and project-level guardrail function-scope filtering. in: header name: X-Opper-Name schema: type: string - description: Parent span ID for distributed tracing context. in: header name: X-Opper-Parent-Span-Id schema: format: uuid type: string - description: >- Comma-separated `key:value` usage-attribution tags (e.g. `tenant:acme,project:demo`, max 8). Recorded on the generation's billing/metrics rows; group spend by any key via GET /v2/analytics/usage?group_by=. Header-borne twin of the /v3/session URL prefix tags (which win per key when both are present); `opper.`-prefixed keys and `session_id` are reserved. Malformed values return 400. in: header name: X-Opper-Tags schema: type: string requestBody: content: application/json: schema: $ref: '#/components/schemas/DecisionRequest' required: true responses: '200': content: application/json: schema: $ref: '#/components/schemas/DecisionResponse' description: Successful response headers: X-Generation-Cost: description: Execution cost in USD, matching X-Opper-Cost. schema: format: double type: number X-Generation-Id: description: Opper generation ID for this evaluation. schema: type: string X-Opper-Cost: description: Execution cost of the request as a floating-point number. schema: format: double type: number X-Opper-Trace-Id: description: Trace UUID for this evaluation. schema: format: uuid type: string '400': content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' description: Bad request '401': description: Unauthorized - missing or invalid API key '500': content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' description: Internal server error components: schemas: DecisionRequest: additionalProperties: false properties: input: oneOf: - type: string - items: properties: content: oneOf: - type: string - items: oneOf: - properties: text: type: string type: enum: - input_text type: string required: - type - text type: object - properties: detail: description: Image detail level. Defaults to auto. enum: - low - high - auto - original type: string image_url: description: >- Inline base64 data URL. Hosted URLs and file IDs are not accepted. pattern: '^data:' type: string type: enum: - input_image type: string required: - type - image_url type: object minItems: 1 type: array role: enum: - user type: string type: enum: - message type: string required: - role - content type: object minItems: 1 type: array model: example: gpt-6-luna minLength: 1 type: string questions: items: $ref: '#/components/schemas/DecisionQuestion' minItems: 1 type: array safety_identifier: description: >- Accepted for compatibility with OpenAI's API, then dropped: Opper neither forwards nor stores it. Requests to OpenAI carry Opper's own identifier for your project. type: string required: - model - input - questions type: object DecisionResponse: additionalProperties: true properties: answers: description: One answer per question, in question order. items: $ref: '#/components/schemas/DecisionAnswer' type: array model: type: string usage: $ref: '#/components/schemas/DecisionUsage' required: - model - answers - usage type: object ErrorResponse: properties: error: properties: code: type: string details: description: Any value message: type: string required: - code - message type: object meta: type: object required: - error type: object DecisionQuestion: oneOf: - additionalProperties: false properties: instructions: type: string name: description: Unique within the request; echoed on the answer. type: string type: enum: - predicate type: string required: - type - instructions type: object - additionalProperties: false properties: choices: items: additionalProperties: false properties: description: type: string value: description: >- A string or a boolean. Values are typed: the string "true" and the boolean true are different choices. type: - string - boolean required: - value type: object maxItems: 255 minItems: 2 type: array instructions: type: string name: description: Unique within the request; echoed on the answer. type: string type: enum: - choice type: string required: - type - instructions - choices type: object - additionalProperties: false properties: instructions: type: string levels: description: Ordered from lowest to highest; level indices start at 0. items: additionalProperties: false properties: description: type: string label: type: string required: - label type: object maxItems: 10 minItems: 2 type: array name: description: Unique within the request; echoed on the answer. type: string type: enum: - score type: string required: - type - instructions - levels type: object DecisionAnswer: oneOf: - additionalProperties: true properties: name: type: - string - 'null' probability: maximum: 1 minimum: 0 type: number type: enum: - predicate type: string required: - type - probability type: object - additionalProperties: true properties: choice: description: >- A string or a boolean. Values are typed: the string "true" and the boolean true are different choices. type: - string - boolean confidence: maximum: 1 minimum: 0 type: number name: type: - string - 'null' probabilities: items: properties: label: type: string probability: maximum: 1 minimum: 0 type: number value: description: >- A string or a boolean. Values are typed: the string "true" and the boolean true are different choices. type: - string - boolean required: - value - probability type: object type: array type: enum: - choice type: string required: - type - choice - confidence - probabilities type: object - additionalProperties: true properties: confidence: maximum: 1 minimum: 0 type: number name: type: - string - 'null' probabilities: items: properties: label: type: string probability: maximum: 1 minimum: 0 type: number value: minimum: 0 type: integer required: - value - probability type: object type: array score: description: Probability-weighted average of the level indices. maximum: 9 minimum: 0 type: number type: enum: - score type: string required: - type - score - confidence - probabilities type: object - additionalProperties: true properties: name: type: - string - 'null' type: enum: - refusal type: string required: - type type: object DecisionUsage: additionalProperties: true properties: input_tokens: minimum: 0 type: integer output_tokens: minimum: 0 type: integer total_tokens: minimum: 0 type: integer required: - input_tokens type: object securitySchemes: BearerAuth: bearerFormat: API Key description: API key authentication. Pass your API key as a Bearer token. scheme: bearer type: http ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.