Getting started

Cheaper Gen. API documentation

One REST API for official video and image generation models from ByteDance, Alibaba and MiniMax — priced below the official list. Submit a job, then poll it or receive a webhook. Switching models means changing one string.

Quickstart

API key to first video in two minutes, with cURL, Python or JavaScript.

Authentication

Bearer API keys, where to keep them and how to rotate them.

Generations

The async lifecycle: queued → running → succeeded or failed.

Webhooks

Get notified when a job finishes and verify the signature.

API reference

Every endpoint, parameter and response object.

Models

All model endpoint IDs with prices and per-model API docs.

Base URL#

All endpoints live under a single versioned base URL. Requests and responses are JSON over HTTPS; amounts are in US dollars.

https://cheapergen.com/api/v1

How it works#

Submit a generation with POST /api/v1/run/{endpoint}. The price is held from your prepaid balance and you get back a generation object with status: "queued" (HTTP 202).
Wait for it to finish: poll GET /api/v1/generations/{id} every few seconds, or pass a webhook_url and we'll POST the result to you. See Generations.
Download the output from output.video.url or output.images[].url. If the job fails, the hold is refunded automatically — you never pay for failed generations.

POST /api/v1/run/bytedance/seedance-2.0-fast/text-to-video

curl -X POST https://cheapergen.com/api/v1/run/bytedance/seedance-2.0-fast/text-to-video \
  -H "Authorization: Bearer $CHEAPERGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "prompt": "A slow dolly-in on a ceramicist shaping a vase at golden hour, dust in the light, shallow depth of field",
  "aspect_ratio": "16:9",
  "resolution": "720p",
  "duration": 5
}'

Models#

Every model exposes one endpoint per task (for example text-to-video and image-to-video), and every endpoint takes the same request shape: the model's input fields plus an optional webhook_url. Browse the full list with prices on the Models page, or fetch it from GET /api/v1/models.

Video: Seedance 2.0, Seedance 2.5, Seedance 2.0 Fast, Seedance 2.0 Mini, Wan 3.0, Wan 3.0 Prime, MiniMax H3, MiniMax H3 Max
Image: Seedream 5.0 Pro, Seedream 5.0 Lite, Seedream 5.0 Flash

Conventions#

Authenticate with Authorization: Bearer cg_live_… — see Authentication.
Generation IDs start with gen_. Timestamps are ISO 8601 in UTC. Money fields end in _usd and are decimal US dollars.
Errors always have the shape { "error": { "code", "message", "param" } } with a meaningful HTTP status — see Errors.
Unknown request parameters are rejected with 400 invalid_request, so typos never silently change your output.
Public endpoints (/models, /quote) need no key and are safe to call from a browser.

For AI agents and LLMs#

A machine-readable summary of the API and every model is available at /llms.txt. Each model also has an API page with its full input schema, generated from the same catalog the API validates against.

Agents can also call Cheaper Gen. directly: connect our hosted MCP server to Claude Code, Cursor or VS Code, or install the Cheaper Gen. agent skill.

Something unclear or missing? Email support@cheapergen.com — we read everything.