> ## Documentation Index > Fetch the complete documentation index at: https://docs.tokenlab.sh/llms.txt > Use this file to discover all available pages before exploring further. # Evaluate Decisions > Evaluate named probability, choice, and score questions with a native decision model. Send `state` and named `questions` to a decision model. Use `GET /v1/models?category=decision` for current availability and `GET /v1/models/{model}` for its published contract. `state` and each question's `instructions` accept a string, JSON object, or JSON array. The endpoint returns synchronous JSON. | Question type | Criteria | Result | | ------------- | --------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- | | `noul` | Optional object with both `true` and `false` guidance | `noul`, a probability from 0 to 1 | | `choice` | Object of 1–255 option names mapped to guidance or `null` | `choice`, optionally `probabilities` and `confidence` | | `score` | Ordered array of 2–10 descriptions | `score` from 0 to the last index; it can be fractional, with optional `legend`, `probabilities`, and `confidence` | The response preserves your question names under `answers`. `usage.input_tokens` and `usage.output_tokens` report observed usage when available. Pricing comes from the model's current tariff; free output does not imply zero output tokens. Chat messages, tools, streaming, and Batch are not part of this endpoint's contract. ```bash theme={null} curl https://api.tokenlab.sh/v1/systemone \ -H "Authorization: Bearer $TOKENLAB_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "jev-1.13", "state": "I was charged twice. Please refund the duplicate payment.", "questions": { "refund_requested": {"type": "noul", "instructions": "Is a refund requested?"}, "department": { "type": "choice", "instructions": "Which team should handle this?", "criteria": {"billing": "Charges and refunds", "technical": "Software bugs"} }, "urgency": { "type": "score", "instructions": "Rate urgency", "criteria": ["Routine enquiry", "Money affected", "Safety emergency"] } } }' ``` The HTTP request is also usable with standard `fetch` or `requests` clients. A decision response is structured data, not a generated text explanation. ## OpenAPI ````yaml openapi.json POST /v1/systemone openapi: 3.1.0 info: title: TokenLab AI Gateway description: >- Organization balance, API key management, and key-level usage/billing via management token version: 1.0.0 termsOfService: https://tokenlab.sh/tos contact: name: Technical Support email: support@tokenlab.sh servers: - url: https://api.tokenlab.sh description: Production server security: - BearerAuth: [] tags: - name: Chat description: Chat completions API (OpenAI-compatible) - name: Responses description: OpenAI Responses API-compatible native endpoints - name: Embeddings description: Text embeddings API - name: Images description: Image generation API - name: Audio description: Audio processing API (TTS & STT) - name: Video description: Video generation API - name: Models description: Available models listing - name: Anthropic description: Anthropic-compatible Messages API - name: Gemini description: Google Gemini-compatible API - name: Management description: >- Organization API key management and key-level usage/billing via management token - name: Files description: Batch file upload and retrieval - name: Batches description: OpenAI-compatible asynchronous batch jobs - name: Seedance Volc Compatible description: Seedance 2.0 Volc-style compatibility endpoints - name: Decisions - name: Webhooks description: >- Workspace webhook management with Management Tokens (mt-...). Inference API keys do not grant management access. paths: /v1/systemone: post: tags: - Decisions summary: Evaluate decision questions description: >- Evaluate state with named probability, choice, and ordered score questions. Synchronous JSON only. Token usage is reported when available; zero-priced output tokens are still included. operationId: createSystemOneDecision parameters: - $ref: '#/components/parameters/DeliveryPolicy' requestBody: required: true content: application/json: schema: type: object required: - model - state - questions properties: model: type: string minLength: 1 state: oneOf: - type: string - type: object additionalProperties: true - type: array items: {} questions: type: object minProperties: 1 additionalProperties: oneOf: - type: object required: - type - instructions properties: type: type: string enum: - noul instructions: oneOf: - type: string - type: object additionalProperties: true - type: array items: {} criteria: type: object properties: 'true': oneOf: - type: string - type: object additionalProperties: true - type: array items: {} 'false': oneOf: - type: string - type: object additionalProperties: true - type: array items: {} required: - 'true' - 'false' additionalProperties: false additionalProperties: false - type: object required: - type - instructions - criteria properties: type: type: string enum: - choice instructions: oneOf: - type: string - type: object additionalProperties: true - type: array items: {} criteria: type: object minProperties: 1 maxProperties: 255 additionalProperties: oneOf: - type: string - type: object additionalProperties: true - type: array items: {} - type: 'null' additionalProperties: false - type: object required: - type - instructions - criteria properties: type: type: string enum: - score instructions: oneOf: - type: string - type: object additionalProperties: true - type: array items: {} criteria: type: array minItems: 2 maxItems: 10 items: oneOf: - type: string - type: object additionalProperties: true - type: array items: {} additionalProperties: false additionalProperties: false responses: '200': description: Decision results content: application/json: schema: type: object required: - model - answers properties: id: type: string model: type: string answers: type: object additionalProperties: oneOf: - type: object required: - type - noul properties: type: type: string enum: - noul noul: type: number minimum: 0 maximum: 1 additionalProperties: false - type: object required: - type - choice properties: type: type: string enum: - choice choice: type: string probabilities: type: object additionalProperties: type: number minimum: 0 maximum: 1 confidence: type: number minimum: 0 maximum: 1 additionalProperties: false - type: object required: - type - score properties: type: type: string enum: - score score: type: number minimum: 0 probabilities: type: object additionalProperties: type: number minimum: 0 maximum: 1 confidence: type: number minimum: 0 maximum: 1 legend: type: object additionalProperties: oneOf: - type: string - type: object additionalProperties: true - type: array items: {} additionalProperties: false usage: type: object required: - input_tokens - output_tokens properties: input_tokens: type: integer minimum: 0 output_tokens: type: integer minimum: 0 additionalProperties: false additionalProperties: true '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '503': $ref: '#/components/responses/DeliveryUnavailable' components: parameters: DeliveryPolicy: name: X-TokenLab-Delivery-Policy in: header required: false description: >- Per-request Delivery policy. Overrides the API key and Workspace defaults. Auto tries TokenLab Verified first and may switch once to Official only before output, upstream acceptance, or persistent resource creation. schema: type: string enum: - auto - verified - official responses: BadRequest: description: Bad Request - Invalid input content: application/json: schema: $ref: '#/components/schemas/ValidationError' example: error: message: 'model: Model is required' type: invalid_request_error param: model validation_errors: - field: model message: Model is required Unauthorized: description: Unauthorized - Invalid or missing API key content: application/json: schema: $ref: '#/components/schemas/ApiError' example: error: message: Invalid API key provided type: invalid_api_key DeliveryUnavailable: description: >- No eligible Delivery route is currently available, or all eligible routes failed. content: application/json: schema: $ref: '#/components/schemas/ApiError' examples: requestedTierUnavailable: summary: The requested Delivery tier has no eligible route value: error: message: >- The requested Delivery tier is temporarily unavailable for model gpt-5.4. type: all_channels_failed code: delivery_tier_unavailable retryable: true request_id: req_01JEXAMPLE eligibleRoutesFailed: summary: All eligible Delivery routes failed value: error: message: All available routes failed to process the request. type: all_channels_failed retryable: true request_id: req_01JEXAMPLE schemas: ValidationError: allOf: - $ref: '#/components/schemas/ApiError' - type: object properties: validation_errors: type: array items: type: object properties: field: type: string message: type: string code: type: string ApiError: type: object required: - error properties: error: type: object properties: message: type: string description: Error message type: type: string description: Error type enum: - unauthorized - invalid_api_key - expired_api_key - permission_error - insufficient_balance - quota_exceeded - invalid_request_error - model_not_found - context_length_exceeded - unsupported_tool_choice - rate_limit_exceeded - server_error - upstream_error - all_channels_failed - timeout_error - not_found_error - no_contract_compatible_route - request_shape_channel_mismatch - upstream_contract_mismatch - platform_normalization_error - async_task_not_found - async_task_mapping_invalid - delivery_tier_unavailable code: type: string description: Error code param: type: string description: Parameter that caused the error model: type: string description: Model associated with the error did_you_mean: type: string suggestions: type: array items: type: object required: - id properties: id: type: string additionalProperties: true alternatives: type: array items: type: object required: - id - status - tags properties: id: type: string status: type: string tags: type: array items: type: string additionalProperties: true hint: type: string retry_after: type: number retryable: type: boolean balance_usd: type: number estimated_cost_usd: type: number supported_operations: type: array items: type: string supported_parameters: type: array items: type: string required_selectors: type: array items: type: string optional_selectors: type: array items: type: string allowed_resolutions: type: array items: type: string allowed_durations: type: array items: type: string allowed_aspect_ratios: type: array items: type: string prompt_max_characters: type: number recommended_request: type: object additionalProperties: true request_endpoint: type: - string - 'null' request_shape_mode: type: - string - 'null' status_mode: type: - string - 'null' request_id: type: string required: - message - type additionalProperties: true additionalProperties: true securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: API Key description: >- API Key authentication. Create or manage API keys in [Dashboard > API > API Keys](https://tokenlab.sh/dashboard/api?tab=keys). ````