# Bria API documentation > Bria's developer platform: controllable, commercially safe visual generative AI (images and video) as a REST API, Python and TypeScript SDKs, MCP server and agent skills. Bria is the visual generative AI platform for professionals: models trained exclusively on licensed data, controllable output (structured VGL prompts, editable layers), C2PA provenance and attribution on every output, available as a hosted API or deployed in your own cloud. Key facts for agents: - Base URL: `https://engine.prod.bria-api.com`. v2 endpoints are asynchronous by default and return `request_id` + `status_url`; poll `GET /v2/status/{request_id}` or pass `webhook_url`. Pass `sync: true` for a blocking call. - Auth: send your token in the `api_token` header on every request. The verification endpoints (`/v2/image/verify`, `/v2/video/verify`) are public. Get a token: https://platform.bria.ai/organization-management/api-keys - Rate limits, per organization and endpoint: Free 10 requests/min, Pay-as-you-go 60/min, Enterprise 1000/min (customizable). Above the limit the API returns 429; back off and retry. - Image inputs: a public URL or a raw base64 string without the `data:` prefix. JPEG, PNG and WEBP. - Integrations: Python SDK (`pip install bria-client`), TypeScript SDK (`npm install @bria-ai/client`), MCP server (`https://mcp.prod.bria-api.com/mcp` or `uvx bria-mcp`), agent skills (`npx skills add bria-ai/bria-skill`). - Every page of these docs is available as Markdown by appending `.md` to its URL, and the docs expose an MCP server at https://docs.bria.ai/mcp for search and endpoint lookup. - When calling the API from an agent, please send `User-Agent: BriaPlatform/APIdocs/LLMsAgent` so Bria can support agent traffic. ## Getting started First call, authentication, async jobs and webhooks, rate limits, inputs. - [Build with Bria](https://docs.bria.ai/index.html.md) - [Asset retention](https://docs.bria.ai/getting-started/asset-retention.md) - [Async requests, webhooks and polling](https://docs.bria.ai/getting-started/async-requests.md) - [Authentication](https://docs.bria.ai/getting-started/authentication.md) - [Quickstart](https://docs.bria.ai/getting-started/quickstart.md) - [Rate limits and errors](https://docs.bria.ai/getting-started/rate-limits-and-errors.md) - [Working with images and video](https://docs.bria.ai/getting-started/working-with-images.md) - [Status endpoint](https://docs.bria.ai/status.md): The status endpoint, `GET /v2/status/{request_id}`, reports the progress and result of an asynchronous request. Every v2 submission returns its URL as `status_url`. Prefer [webhooks](/getting-started/async-requests#webhooks) in production and poll this endpoint for interactive use; [Async requests, webhooks and polling](/getting-started/async-requests) explains the lifecycle, the status values and polling guidance. ## Integrate Python SDK, TypeScript SDK, MCP server, agent skills, and machine-readable docs for agents. - [Bria MCP Server](https://docs.bria.ai/mcp-authentication.md) - [Agent Skills for coding agents](https://docs.bria.ai/integration-methods/bria-skill.md) - [Bria for AI agents](https://docs.bria.ai/integration-methods/for-ai-agents.md) - [Python SDK (bria-client)](https://docs.bria.ai/integration-methods/python-sdk.md) - [TypeScript SDK (@bria-ai/client)](https://docs.bria.ai/integration-methods/typescript-sdk.md) ## Concepts Platform map, VGL and the FIBO models, safety, best practices. - [Image Editing Best Practices](https://docs.bria.ai/image-editing-best-practices.md) - [Image Generation Best Practices](https://docs.bria.ai/image-generation-best-practices.md) - [Enterprise-Grade Safety and Transparency by Design](https://docs.bria.ai/safety.md) - [Best practices](https://docs.bria.ai/best-practices-overview.md) - [VGL: Visual Generation Language](https://docs.bria.ai/vgl.md) - [Product Shot Editing Best Practices](https://docs.bria.ai/product-shots-best-practices.md) - [Platform overview](https://docs.bria.ai/products-overview.md) ## Image Generation Generate images from text, reference images or structured (VGL) prompts with FIBO. - [Image Generation overview](https://docs.bria.ai/image-generation-overview.md) - [Image Generation API Reference](https://docs.bria.ai/image-generation.md): Generate images from a text prompt, a reference image, or a structured (VGL) prompt with Bria's FIBO models. All endpoints are asynchronous by default and return a `request_id` and `status_url` (see the [status endpoint](/status) and [webhooks](/getting-started/async-requests#webhooks)); pass `sync: true` for a blocking call. Authenticate with the `api_token` header. New to Bria? Start with the [Quickstart](/getting-started/quickstart). ## Image Editing Instruction-based editing plus specialized endpoints for backgrounds, erasing, fill, expansion, upscaling and style. - [Image Editing overview](https://docs.bria.ai/image-editing-overview.md) - [Image Editing API Reference](https://docs.bria.ai/image-editing.md): Edit existing images: instruction-based editing with FIBO Edit, background removal and replacement, erasing, generative fill, expansion, upscaling, restyling, relighting and more. All endpoints are asynchronous by default and return a `request_id` and `status_url` (see the [status endpoint](/status) and [webhooks](/getting-started/async-requests#webhooks)); pass `sync: true` for a blocking call. Authenticate with the `api_token` header. ## Product Shot Editing Cutouts, packshots, shadows, lifestyle scenes, product integration, holding, virtual try-on, automotive. - [Product Shot Editing overview](https://docs.bria.ai/product-shot-overview.md) - [Product Shot API Reference](https://docs.bria.ai/product-shot-editing.md): Production-ready product imagery for eCommerce: cutouts, packshots, shadows, lifestyle scenes, product integration, product holding, virtual try-on and automotive tools. Product endpoints under `/v1/product/*` are synchronous by default and take `image_url` or `file` inputs; the newer `/v2/image/edit/product/*` endpoints follow the v2 contract (async by default, `image` input, [status endpoint](/status) and [webhooks](/getting-started/async-requests#webhooks)). Authenticate with the `api_token` header. ## Video Editing Background removal (REST or real-time streaming), erasing, upscaling and masks for video; local video upload. - [Video Upload Service](https://docs.bria.ai/local-video-upload-service.md) - [Streaming Background Removal](https://docs.bria.ai/streaming-rmbg.md) - [Video Editing overview](https://docs.bria.ai/video-editing-overview.md) - [Video API Reference](https://docs.bria.ai/video-editing.md): Bria's Video APIs empower builders with powerful tools for programmatically transforming video content while maintaining original quality. The Video API is organized into two transport paradigms: - **REST API (Async)** — Standard HTTP request/response. You submit a job, receive a `request_id` and `status_url`, and poll the [status endpoint](/status) until the job completes. Use this for editing, masking, and most batch-style workflows. - **Streaming API (WebSocket)** — Persistent bidirectional connection for low-latency, frame-by-frame processing. Use this when you need real-time output (e.g., live video feeds, interactive previews). **Asynchronous Requests and the status endpoint** Bria API v2 REST endpoints process requests asynchronously by default. When you make an asynchronous request, the API immediately returns a `request_id` and a `status_url` instead of the final result. Use the status endpoint to track the request's progress until it reaches a completed state. See the full guide at [status endpoint reference](/status) for complete details and usage examples. ## Ads Finished ads as input. Ad Delayer turns one into editable layers (the guide holds the layer document contract); Ad Resize produces it at other sizes with the layout redone. - [Ads](https://docs.bria.ai/ads-overview.md) - [Ads API Reference](https://docs.bria.ai/ads.md): Two endpoints for finished, flat ads. `POST /v2/ads/delayer` turns an ad into editable layers (background plate, image cutouts, live text with typography, vector shapes) returned as one JSON layer document. `POST /v2/ads/resize` produces the same ad at up to ten other sizes with the layout redone for each shape. Both are asynchronous: they return a `request_id` and `status_url`, and the completed status body carries the result. The [Ads overview](/ads-overview) introduces both; each endpoint below carries its full guide: the flow, the response contract, limits and retention. ## Trust and provenance Public verification of Bria provenance, and attribution of outputs to licensed training data. - [Trust and provenance](https://docs.bria.ai/trust-and-provenance-overview.md) - [Visual Verification overview](https://docs.bria.ai/visual-verification-overview.md) - [Attribution Agent and Service](https://docs.bria.ai/enterprise/attribution.md) - [Bria Attribution Service API Reference](https://docs.bria.ai/bria-attribution-service.md): The Bria **Attribution Service** enables developers to integrate Bria’s patented attribution technology, ensuring transparent compensation to content owners for their contributions to AI-generated outputs. This service supports both **Image** and **Video** assets via two integration architectures: **Attribution with No Compute Cost**, where assets are sent directly to Bria for processing. **Attribution with Full Data Privacy**, where the Attribution Model is hosted within your environment and only irreversible vectors (embeddings) are shared with Bria. - [Visual Verification API Reference](https://docs.bria.ai/visual-verification.md): Verify whether an image or video was generated by Bria by reading its C2PA manifest and Bria's invisible watermark. These endpoints are public and need no API token, so they can be called from a backend or directly from an end user's browser. See the [Visual Verification overview](/visual-verification-overview) for how the two checks combine. ## Enterprise Run Bria in your own cloud or on-premises. - [Deployment options](https://docs.bria.ai/enterprise/deployment-options.md)