Getting Started

📝 TODO — this English page is a skeleton. The structure mirrors the Chinese page, which is fully written. Contributions and translations go here.

ModelFold exposes video, image, chat and speech models behind one OpenAI-style HTTP API. With a single API key you can call any of the 192 models.

1. Create an account

Register in the ModelFold console and verify your email address. New accounts start with a credit balance for testing.

2. Create an API key

Open the API Keys section of the console and create a key. Keys look like this:

mf-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

• A single account can hold up to 10 active keys.
• Keys can be viewed, copied and revoked at any time; revocation takes effect immediately (401 afterwards).

3. Send your first request

export API_BASE="https://ai.modelfold.online/v1"   # becomes https://ai.modelfold.online/v1 once the domain is live
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"}'

4. List available models

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

5. Understand credits

• Balances are checked before the upstream is ever called; an insufficient balance returns 402 without spending anything.
• Media (image / video / audio) is charged per call.
• Chat is settled from the real token usage the upstream reports — pre-authorized, then adjusted up or down.
• Failed upstream calls are refunded automatically and reported as 502.

Next steps

• Authentication
• Your first request
• API Reference