The multimodal API layer for developers

Access video, image, music and LLM models through one API.

API-Key lets customers try models in the browser, create API keys, and ship the same model capabilities to their own users with a managed gateway, credit wallet and task system.

Start buildingRead API docs

One account. One wallet. One API key. Browser playgrounds and production endpoints share the same model gateway.

API request

Async media task · managed gateway

live gateway
curl https://api.api-key.org/api/v1/jobs/createTask \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "bytedance/seedance-2-fast",
    "input": {
      "prompt": "A cinematic product launch video with clean motion",
      "duration": 5
    },
    "callBackUrl": "https://your-app.com/webhooks/api-key"
  }'

task_id

returned instantly

credits

reserved then settled

result link

durable media URL

One API

Video, image, music and LLM access

Tasks

Async status, polling and callback support

0 credits

Refund policy for failed model tasks

120+

LLM, image, video and audio models

Browser playground

Let customers try models before they call the API.

The web console stays connected to the same credit system and media workflows as the developer API. A workflow that works here can be exposed to end users through API keys.

Generate and edit commercial visuals.

Session historyImages generated in this session will appear here.

Featured models

A model marketplace for media and LLM APIs.

View endpoint reference →
Videofrom credits / second

Seedance 2

Fast text-to-video, image-to-video and video-to-video generation for product demos, ads and cinematic clips.

Try model →

Videotask-based credits

Grok Imagine

Create short motion clips from text or reference images with an API-ready task workflow.

Try model →

Imageper image credits

GPT Image 2

High-quality image generation and editing with stronger text rendering, product scenes and ad creative outputs.

Try model →

Imageper image credits

Nano Banana 2

Fast image generation and editing for creators who need consistent characters, references and rapid iteration.

Try model →

Musicper task credits

Music generation

Generate tracks from prompts and expose them through the same task API surface as visual media.

Try model →

LLMper token

Claude, GPT, Gemini and Grok

Messages, Responses and Chat Completions endpoints with per-token billing, prompt caching and streaming.

Try model →

Developer API

Turn every model into a product-ready API for your customers.

Consumers can generate keys, call compatible routes, submit async media tasks, poll results, track usage and pay from one credit balance. The gateway turns model capabilities into a developer product.

Open API documentationManage API keys
curl https://api.api-key.org/v1/messages \
  -H "x-api-key: $API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "max_tokens": 1024,
    "messages": [
      { "role": "user", "content": "Create a launch plan" }
    ]
  }'
GET/v1/models

List public models and categories available through API-Key.

POST/v1/messages

Anthropic Messages API for Claude models. Works with Claude Code.

POST/v1/responses

OpenAI Responses API for GPT, Codex and Grok models.

POST/api/v1/jobs/createTask

Create an async task for any of 80+ image, video and audio models.

POST/v1/videos

OpenAI-style alias for video tasks; image and music aliases are available too.

GET/v1/tasks/{task_id}

Poll async image, video and music task status and output URLs.

Why managed model routing matters

Direct integrations become a managed API platform.

API-Key adds the product controls needed to sell model access: keys, channels, model groups, retries, billing settlement and logs.

Use every model from one wallet

Give developers one account, one credit balance, one API key and one usage history across text, image, video and music generation.

Turn playground output into production APIs

Customers can test models inside API-Key, save their best workflows, then call the same capabilities from their own products.

Native protocols for every LLM

Claude on the Messages API, GPT and Grok on the Responses API, Gemini on Chat Completions. Point existing SDKs and Claude Code at one base URL.

Store generated media for real apps

Async image, video and music tasks return trackable records and durable result links instead of short-lived temporary URLs only.

From demo to API revenue

The same capability works for your UI and your customers' apps.

01

Try a model in the browser

Use the visual playground for images, video and music, or call LLMs with the API examples.

02

Create an API key

Generate keys with model limits, IP allowlists, rate limits and usage tracking.

03

Hold, settle, refund

Each call reserves credits up front and settles to the real token or task cost. Failed tasks are refunded in full.

04

Bill and deliver results

Reserve credits, refund failed tasks, expose durable result links and task records to your product.

FAQ

Built for developers who want to sell AI capabilities.

Build your model API business with API-Key.

Start with the browser playground, connect your upstream channels, publish API keys and let your own customers generate text, images, video and music through API-Key.

Create accountView API reference