OyeFilmy Logo
personFreeUpgrade
bolt0cr
U

Tools

SDKs

Official SDK libraries for Python and Node.js are in development.

Not published yet. pip install oyefilmy and npm install @oyefilmy/sdk do not resolve — neither package is on PyPI or npm today. The snippets below show the intended interface, not something you can install right now. Until they ship, call the REST API directly: see the Quickstart, which uses nothing but fetch and requests.

Generate a client from the OpenAPI spec

You do not have to wait for us. The API publishes an OpenAPI 3.1 document, generated from the same model routing tables the server dispatches on — so the model keys, duration bounds and scopes in it are the live ones, not a copy that drifted. Point any generator at it:

# Inspect it
curl https://api.oyefilmy.com/api/openapi.json

# Python client
openapi-python-client generate --url https://api.oyefilmy.com/api/openapi.json

# TypeScript types
npx openapi-typescript https://api.oyefilmy.com/api/openapi.json -o oyefilmy.d.ts

A self-hosted install serves its own origin in the servers block, so a client generated from your deployment points at your deployment. If your tooling speaks the Model Context Protocol, skip code generation entirely and use the MCP server— same capability, no client to maintain.

Planned installation

Python

pip install oyefilmy

Node.js

npm install @oyefilmy/sdk

Usage

# Install
pip install oyefilmy

# Usage
from oyefilmy import OyeFilmy

client = OyeFilmy(api_key="of_live_YOUR_API_KEY")

# Generate an image
result = client.generate(
    model="gpt",
    prompt="A cinematic wide shot of a vintage red scooter on a rainy Mumbai street",
    aspect_ratio="16:9",
)
print(f"Image URL: {result.output_url}")

# Generate a video
result = client.generate(
    model="seedance_2_0_fast_t2v",
    prompt="A drone shot flying over misty mountains at sunrise",
    duration=5,
)
print(f"Video URL: {result.output_url}")

# Check balance
balance = client.credits.balance()
print(f"Credits: {balance.balance}")

# With webhook (fire and forget)
job = client.generate(
    model="gpt",
    prompt="A watercolor painting of a sunset",
    webhook_url="https://yourapp.com/webhook",
    wait=False,  # Don't poll — webhook will deliver
)
print(f"Job ID: {job.prediction_id}")

SDK features

FeaturePythonNode.js
Image generation
Video generation
Audio generation
3D generation
Automatic polling
Webhook support— (API only, verify it yourself)— (API only, verify it yourself)
Voice clone
Image upscale
Credit balance
Generation history
TypeScript types—
Async/await (asyncio)
Retry logic

Planned behaviour: both SDKs will handle polling automatically — generate() blocks until the result is ready, with wait=False (Python) or wait: false (Node.js) to get the prediction ID immediately. Neither package is published yet.