Getting Started

ModelFold brings video, image, chat and speech models together behind an OpenAI-style HTTP API. A single API key gives you access to any of the 186 models.

1. Create an account

Registration uses POST /api/auth/register. Passwords must contain at least eight characters, including both letters and digits. See Authentication API.

2. Create an API key

text

mf-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

An account can have up to 10 active keys at a time.
You can view or revoke a key in the console at any time. Revocation takes effect immediately; requests using the revoked key return 401.

3. Send your first request

All model requests use /v1. Pass your key in the Authorization: Bearer header:

bash

export API_BASE="https://modelfold.ai/v1"   # Production API base URL
export MODELFOLD_API_KEY="mf-..."

curl -X POST "$API_BASE/images/generations" \
  -H "Authorization: Bearer $MODELFOLD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nano-banana-2","prompt":"a paper boat drifting down a rainy street"}'

Responses follow the { ok, created, data, usage } structure. data[0].url is the output URL, and usage includes the credits charged and your remaining balance.

4. List available models

bash

curl "$API_BASE/models" -H "Authorization: Bearer $MODELFOLD_API_KEY"

You can also browse the model catalog. Each model has its own reference page.

5. Understand credits

The gateway checks your balance before each call. If it is too low, the request returns 402 without calling the model provider.
Media models (image, video and audio) are charged per call.
Chat models are billed from actual token usage: credits are reserved based on an estimate, then the difference is refunded or charged when usage is available.
Failed provider calls are refunded automatically. The response returns 502 with the failure reason.

See Pricing for billing details.

Getting Started ​

1. Create an account ​

2. Create an API key ​

3. Send your first request ​

4. List available models ​

5. Understand credits ​