API-Key API Documentation

Base URL and authentication

https://api.api-key.org

Authorization: Bearer YOUR_API_KEY
# or
x-api-key: YOUR_API_KEY

Create keys at Settings → API Keys. Never ship a key in browser or mobile code.

Billing

Rule	Details
Unit	1 credit = $0.001. Prices for every model are listed at /pricing and in `GET /v1/models`.
LLM	Billed per input, output and cached token. A hold based on `max_tokens` is placed before the call and settled to actual usage.
Media tasks	A hold is placed when the task is created and settled to the actual cost when it finishes. Failed tasks are refunded in full.
Balance	`GET /v1/credits` returns your balance in credits and USD.

LLM endpoints

POST

/v1/messages

Anthropic Messages API for Claude models (streaming, tools, prompt caching). Works with Claude Code.

POST

/v1/responses

OpenAI Responses API for GPT, Codex, Grok and DeepSeek models.

POST

/v1/chat/completions

OpenAI Chat Completions for Gemini and GPT-5.2 models.

GET

/v1/models

Every model with its endpoint and USD price.

Each model is served on one endpoint; the endpoint field in GET /v1/models tells you which.

Claude Code

export ANTHROPIC_BASE_URL=https://api.api-key.org
export ANTHROPIC_AUTH_TOKEN=YOUR_API_KEY
export ANTHROPIC_MODEL=claude-sonnet-5-5
claude

Messages example

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": "Write a launch tweet" }]
  }'

Responses example

curl https://api.api-key.org/v1/responses \
  -H "Authorization: Bearer $API_KEY" \
  -H "content-type: application/json" \
  -d '{ "model": "gpt-6-sol", "input": "Summarise this changelog", "stream": true }'

Media tasks

POST

/api/v1/jobs/createTask

Create a task for any image, video or audio model: { model, input, callBackUrl }.

GET

/api/v1/jobs/recordInfo?taskId=

Task state (waiting, generating, success, fail), result URLs and credits used.

POST

/v1/images/generations · /v1/videos · /v1/music/generations

OpenAI-style aliases. Pass any model id; extra fields become the task input.

GET

/v1/tasks/{task_id}

OpenAI-style task status with result URLs.

Create a task

curl https://api.api-key.org/api/v1/jobs/createTask \
  -H "Authorization: Bearer $API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "nano-banana-2",
    "input": { "prompt": "A product photo of a ceramic mug", "aspect_ratio": "1:1", "resolution": "1K" },
    "callBackUrl": "https://your-app.com/webhooks/api-key"
  }'

{ "code": 200, "msg": "success", "data": { "taskId": "task_..." } }

Poll the result

curl "https://api.api-key.org/api/v1/jobs/recordInfo?taskId=task_..." \
  -H "Authorization: Bearer $API_KEY"

{
  "code": 200,
  "msg": "success",
  "data": {
    "taskId": "task_...",
    "model": "nano-banana-2",
    "state": "success",
    "resultJson": "{\"resultUrls\":[\"https://api.api-key.org/files/task_.../0\"]}",
    "creditsConsumed": 52
  }
}

Input fields for each model are listed on its card at /models. Result files are kept for 14 days; download anything you need to keep.

Webhooks

When callBackUrl is set, the task record is POSTed to it when the task succeeds or fails. Generate a signing key at Settings → API Keys to receive signed callbacks:

X-Webhook-Timestamp: 1790000000
X-Webhook-Signature: base64(HMAC-SHA256(taskId + "." + timestamp, webhookSigningKey))

Limits

Limit	Default
Requests per API key	5 per 10 seconds
Concurrent media tasks per API key	2
Higher limits	Contact support with your expected volume.

Limited requests return HTTP 429 with a retry-after header.

Errors

{
  "error": {
    "message": "Insufficient credits: this request needs up to 120 credits.",
    "type": "insufficient_quota",
    "code": "insufficient_credits"
  },
  "request_id": "..."
}

/v1/messages returns the Anthropic error shape and /api/v1/jobs/* returns { code, msg, data }.