Model
- GPT Image 2.5 Flare
- Nano Banana 2.1
- GPT Image 2.5 Sunburst
- Seedream 5 Pro
Brief
Teenage EngineeringPortrait Instagram post, “meet OP–XY”
1024 × 15361 Images
Image, video and audioAI Gateway
Create images, audio and video with AI Gateway.
Use leading AI models directly from Rainbrand, with Briefs to guide your images, videos and scripts.
AI Gateway
(01) Why use the AI Gateway
Your agents can make images, video and voiceovers, but every model provider needs its own account and setup. Rainbrand gives your agents a single connection to the best models, guided by your brand's Briefs.
Every format, one key
Generate anything with the frontier AI models.
1024 × 1536Seedance 2.50:05 · 3:2
First frameThe header image
0:00.0
0:01.2
0:03.7
0:04.9
2 Video
Your product images and illustrations, animated into video clips.
Eleven v40:01.1 / 0:06
Script
Meet OP–XY. Same spirit of exploration, a different way to make music.
3 Audio
Turn a script into speech with the voice you choose.
With your agent
Tell your agent what you want to create.
Your agent can request a Brief, choose a model and generate images or video in the same conversation. Use a Review to check what it makes.
Read the docs (opens in a new tab)
ChatGPTYou
Make a 5-second launch teaser of Teenage Engineering's K.O.–SIDEKICK with Seedance 2.5, starting from our product shot.
ChatGPTgenerate_videoK.O.–SIDEKICK video
Rainbrand generationK.O.–SIDEKICK video
Video ready- Model
- Seedance 2.5
- First frame
- A K.O.–SIDEKICK shot
- Length
- 5 seconds, 1178 × 7863:2
The teaser is ready. Seedance kept the shot's 3:2 shape.
Made for this page through Rainbrand's AI Gateway with a Teenage Engineering Brief. Teenage Engineering is a real brand used for illustration, not an endorsement.
In your product
Generate through one API.
Choose a model, send the request and read the output. Include a completed Brief with supported image and video requests to pass in brand guidance and references.
View API reference (opens in a new tab)POST /generations
Generate with a model and a Brief (opens in a new tab)
1Start a generation.
202 Acceptedcurl -X POST https://www.rainbrand.com/api/v1/generations \ -H "x-api-key: $RAINBRAND_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "brandSessionId": "e6220a5e-4a68-4d39-8bb9-5105e9a3a164", "model": "openai/gpt-image-2.5-sunburst", "briefId": "657bcf79-1f2f-47fa-90f8-9245142980ed", "input": { "prompt": "A landscape launch image for Acme Inc. Atlas Pro.", "size": "1536x1024" } }'2Read the output.
200 OKcurl "https://www.rainbrand.com/api/v1/generations/$GENERATION_ID?wait=true" \ -H "x-api-key: $RAINBRAND_API_KEY""data": { "status": "completed", "outputs": [{"url": "https://…"}] }
Customers
Teams building with Rainbrand.
How companies integrate Rainbrand into their agents and products.
All customer stories“We use Rainbrand to fetch each customer’s brand and generate accurate, on-brand creative for their ads. It means we don’t have to reinvent that part of the stack ourselves.”
“Rainbrand is our brand brain for our company and all our assets are generated from there. I don’t know what magic they’ve done but it’s worlds above other generation products and the MCP is delightful to use.”
“I have the ideas, but turning them into great design was always a struggle. Rainbrand makes me feel like the best designer in the world, and I don’t think I’ll need to hire one for the small stuff again.”
Get started with AI Gateway today.
Connect Rainbrand to your agent and ask for an image, video or voiceover, or generate through the API.
Open the Rainbrand plugin and add it.
Open Rainbrand plugin (opens in a new tab)Connect your Rainbrand account when asked. It works in ChatGPT and in Codex.
What can I create with Rainbrand's AI Gateway?
Images, videos and speech, using the models available through Rainbrand. Published examples include product photos, campaign images, animated product shots and voiceovers. Agents can combine those outputs with code and design tools to make pages, packaging and complete campaigns. See the Teenage Engineering run.
Which AI models can I use?
For images: Google's Nano Banana 2.1, OpenAI's GPT Image 2.5 Flare and Sunburst, and ByteDance's Seedream 5 Pro. For video: ByteDance's Seedance 2.5. For speech: ElevenLabs' Eleven v4. Your agent can list the current models and read their options before generating. See how to choose a model.
How do I make AI images and videos on-brand?
Request a Brief and include it with a supported generation request. Rainbrand passes its guidance and visual references to the model alongside your prompt. Your prompt specifies the content and exact wording. Check the result with a Review, using an image or captured video frame, then revise where needed. Read about keeping AI assets on-brand.
Can I use my own images and references?
Yes, where the selected model supports them. You can supply references alongside a Brief, or animate a starting image with a supported video model. The accepted formats, number of references and combinations depend on the model. Follow the product-video example.
Can I make different versions of the same campaign?
Yes. Reuse a relevant Brief and change the prompt, composition or dimensions for each generation. The rabbit example explores three campaign directions, then generates square and landscape versions of the selected one. See the campaign-image guide.
How does audio use our brand voice?
Your agent uses the brand guidance to write the words, then sends the script to a speech model with a selected voice. The brand's tone comes through the writing; the spoken voice is a model setting. Your agent can retrieve the audio for you to listen to.
Do I need separate accounts or API keys for each model?
No. Access the available models through your Rainbrand connection or API key. You can also keep using your own generation tools with a Brief's guidance and assets. The AI Gateway is optional. Use your own generation tools.
Rainbrand builds and maintains shared brand context for your agents and products.
