Image / Video Generation

With a single token, generate images and videos right in your browser — top-tier models like nano-banana, Veo 3.1, Runway, Adobe Firefly, Seedream, Grok, no coding required.

Prerequisite · Pick the right group

Image / video models live in the media / media-gen group. Head to Console → Tokens and create a new token using that group, otherwise you'll get a "No available channel" error. See Choosing a group for details.

Two ways to use it

1 · Online Generation Playground (easiest, no code)

Open the playground, enter your token, and you can generate, download, and view your history:

🎨 Open the Generation Playground →

On the left, enter your token, pick a model, write a prompt, and choose an aspect ratio / resolution / duration (optionally upload reference images — multiple are supported); on the right, your image / video appears instantly.

2 · Call it via the API (integrate into your own program)

Standard OpenAI-style endpoints:

• Images — POST /v1/images/generations (text-to-image), POST /v1/images/edits (image-to-image), returning base64 synchronously
• Video — POST /v1/videos to create a task → poll its status → download the mp4 (asynchronous, Sora-style)

See Calling via the API for details.

What you can generate

• Image generation — 9 models including gpt-image-2, nano-banana-2, seedream-4.5, flux-kontext-max, and the Firefly series; up to 4K; supports image-to-image (multiple reference images)
• Video generation — 5 models including gemini-veo31, runway-gen4-turbo, grok-video, and the Firefly series; supports image-to-video and custom durations (4–15 seconds)

Pricing

Images are billed per image; videos are billed per second (unit price × duration). Each model's live price is shown directly on the "Generate" button in the Generation Playground as you pick a model / resolution / duration.