> ## Documentation Index > Fetch the complete documentation index at: https://docs.prisme.ai/llms.txt > Use this file to discover all available pages before exploring further. # Generate images > OpenAI-compatible image generation. The gateway resolves the model (request `model`, else the org governance default, else the platform default `config.defaults.image_generation`) and routes the call to the provider (OpenAI, Azure OpenAI, Google Vertex Imagen). **Size resolution.** Pass an abstract `aspect` + `tier`, or an explicit pixel `size`. The gateway snaps it to a size listed in the model's `capabilities.supported_sizes`; an exact supported `size` is kept as is. Without a supported-sizes list, `size` is passed through (default `1024x1024`). The size actually used is returned in `size`. **Images are always returned as URLs.** When the provider only returns base64 data, the gateway uploads it to Prisme.ai file storage and returns a tokenised `url` instead, so `data[].b64_json` is never present in the response. **Governance.** Requires the `image-generation` model permission. A model not allowed for the organization is either downgraded to the governance fallback model (`soft_downgrade` quota policy) or refused with `403 MODEL_NOT_ALLOWED`. A required upstream identity header that would reach the provider empty is refused with `403 HEADER_POLICY_VIOLATION`. ## OpenAPI ````yaml /api-reference/llm-gateway/swagger.yml post /v1/images/generations openapi: 3.0.3 info: version: 1.0.0 title: LLM Gateway API description: | Public REST API for the Prisme.ai LLM Gateway - OpenAI-compatible chat completions, embeddings and image generation/editing, a standard rerank endpoint, a managed model catalogue with governance overrides per organization, and quota status for users and organizations. The gateway abstracts multi-provider LLM access (OpenAI, Azure OpenAI, Anthropic, Google Vertex, AWS Bedrock, OpenAI-compatible providers) behind an OpenAI-compatible request/response shape. It enforces per-tenant governance (allowed models, default models, quotas) and emits analytics events (`analytics.llm.completion`) usable for cost and carbon reporting. This spec documents only the public REST surface (endpoints exposed via Prisme.ai workspace webhooks). Internal helpers (private automations prefixed with `_`) and load-test mocks are not part of the public contract. contact: name: Prisme.ai url: https://prisme.ai servers: - url: https://{host}/v2/workspaces/slug:llm-gateway/webhooks description: Prisme.ai workspace webhooks variables: host: default: api.studio.prisme.ai description: API host (override for self-hosted or sandbox) security: - BearerAuth: [] - OrgApiKeyAuth: [] tags: - name: Completions description: OpenAI-compatible chat completions (with optional SSE streaming). - name: Embeddings description: OpenAI-compatible text embeddings. - name: Rerank description: >- Rerank documents by relevance to a query (standard rerank contract, LLM-as-reranker backend). - name: Images description: OpenAI-compatible image generation and image editing. - name: Models description: Model catalogue (CRUD + bulk replace + governance-aware listing). - name: Audio description: OpenAI-compatible speech-to-text (audio transcriptions). - name: Quotas description: >- Effective LLM quota limits and current usage, per user and per organization. - name: Defaults description: >- Resolved default models for completions / embeddings / image generation / transcription / file parsing. - name: Test description: Smoke-test reachability of a model through the gateway. paths: /v1/images/generations: post: tags: - Images summary: Generate images description: | OpenAI-compatible image generation. The gateway resolves the model (request `model`, else the org governance default, else the platform default `config.defaults.image_generation`) and routes the call to the provider (OpenAI, Azure OpenAI, Google Vertex Imagen). **Size resolution.** Pass an abstract `aspect` + `tier`, or an explicit pixel `size`. The gateway snaps it to a size listed in the model's `capabilities.supported_sizes`; an exact supported `size` is kept as is. Without a supported-sizes list, `size` is passed through (default `1024x1024`). The size actually used is returned in `size`. **Images are always returned as URLs.** When the provider only returns base64 data, the gateway uploads it to Prisme.ai file storage and returns a tokenised `url` instead, so `data[].b64_json` is never present in the response. **Governance.** Requires the `image-generation` model permission. A model not allowed for the organization is either downgraded to the governance fallback model (`soft_downgrade` quota policy) or refused with `403 MODEL_NOT_ALLOWED`. A required upstream identity header that would reach the provider empty is refused with `403 HEADER_POLICY_VIOLATION`. operationId: createImage requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ImageGenerationRequest' responses: '200': description: Generated images. content: application/json: schema: $ref: '#/components/schemas/ImagesResponse' '401': description: Missing or invalid authentication. content: application/json: schema: $ref: '#/components/schemas/Error' '403': description: > Missing `image-generation` permission (`FORBIDDEN`), model refused by the organization governance (`MODEL_NOT_ALLOWED`), or header-policy refusal (`HEADER_POLICY_VIOLATION`). content: application/json: schema: $ref: '#/components/schemas/Error' default: description: | Provider error, forwarded with the provider's HTTP status. content: application/json: schema: $ref: '#/components/schemas/Error' components: schemas: ImageGenerationRequest: type: object properties: model: type: string maxLength: 256 description: >- Image model id. Defaults to the org governance default, then the platform default. prompt: type: string description: Text description of the image to generate. size: type: string description: >- Explicit `WxH` pixel size. Kept as is when the model supports it, otherwise snapped (see `aspect` / `tier`). example: 1024x1024 aspect: type: string enum: - square - portrait - landscape description: >- **Prisme.ai extension.** Abstract orientation, snapped to a size the model supports. tier: type: string enum: - small - medium - large description: >- **Prisme.ai extension.** Size tier within the orientation (default `medium`). style: type: string description: Provider style hint, forwarded to the provider when set. 'n': type: number description: Number of images to generate (default 1). response_format: type: string enum: - url - b64_json description: | Format requested from the provider (default `url`). The response always carries URLs: base64 results are uploaded and replaced by a `url`. analytics_context: type: object description: >- **Prisme.ai extension.** Caller-supplied analytics context (`orgSlug`, `agent_id`, `user_id`, `context_id`, `agent_allowed_models`). additionalProperties: true ImagesResponse: type: object properties: created: type: integer description: Unix timestamp of the generation. model: type: string description: Model that produced the images (after governance resolution). data: type: array items: type: object properties: url: type: string description: >- URL of the image (a tokenised Prisme.ai file URL when the provider returned base64). size: type: string description: '**Prisme.ai extension.** Size actually used after resolution.' example: created: 1767225600 model: gpt-image-1 size: 1024x1024 data: - url: https://api.studio.prisme.ai/v2/files/file-abc.png?token=token-123 Error: type: object required: - error description: | Standard error envelope. `error` carries either a stable PascalCase identifier or a free-text label (legacy endpoints) - `code` is the canonical machine-readable identifier going forward. properties: error: type: string description: Stable PascalCase identifier or short error label. message: type: string description: Human-readable error message. code: type: string description: | Machine-readable error code. Observed values include `RATE_LIMITED`, `MODEL_NOT_ALLOWED`, `HEADER_POLICY_VIOLATION`, `MODEL_NOT_FOUND`, `MODEL_EXISTS`, `MISSING_MODEL_ID`, `MISSING_TYPE`, `INVALID_BODY`, `BULK_REPLACE_FAILED`, `INVALID_ITEMS`, `INVALID_DIMENSIONS`, `PAYLOAD_TOO_LARGE`, `METHOD_NOT_ALLOWED`, `PROVIDER_ERROR`, `PROVIDER_NO_RESPONSE`. details: description: Optional structured context (e.g. list of invalid items). additionalProperties: true status: type: integer description: HTTP status mirror, when present. retryAfter: type: integer description: Seconds to wait before retrying (rate-limit responses). provider: type: string description: Upstream provider name (provider-error responses). model: type: string description: Model id involved in the error (provider-error responses). provider_error_type: type: string description: Upstream provider's own error class name. securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: JWT description: | User-bound credential carrying an identity: either a session JWT or a user access token (`at:*`) generated from the user settings UI. Send as `Authorization: Bearer `. Org API keys (`iak_*`) are **not** accepted here - they carry no user identity. Use the `x-prismeai-api-key` header instead (see `OrgApiKeyAuth`). OrgApiKeyAuth: type: apiKey in: header name: x-prismeai-api-key description: | Organization API key (`iak_{orgSlug}_{uuid}`). Unlike `Authorization: Bearer`, this credential is **not** tied to a user identity - it is bound to the org and its effective access is defined by the scopes / permission rules attached to it (it can be restricted to a single project, or kept broader). Send as `x-prismeai-api-key: iak_...`. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.