Back
Developers
The FunAI API
One API for chat, image, video, audio and 3D generation across the leading model makers. The chat and image endpoints are OpenAI-compatible, so the OpenAI SDKs work with a base URL and key change.
| Base URL | https://api.funai.art |
| API keys | Created on the developer portal |
| Billing | A prepaid API balance in USD, separate from FunAI Studio plans |
| Reference | docs.funai.art |
Quick start
curl https://api.funai.art/v1/chat/completions \
-H "Authorization: Bearer $FUNAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-4o-mini", "messages": [{"role": "user", "content": "Hello!"}]}'
from openai import OpenAI
client = OpenAI(base_url="https://api.funai.art/v1", api_key="YOUR_FUNAI_API_KEY")
response = client.chat.completions.create(
model="gpt-4o-mini",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
What you can build with it
- Image generation and editing with FLUX, GPT Image, Nano Banana, Seedream and the rest of the image models.
- Video generation from text or images with Kling, Veo, Seedance, PixVerse and more video models. Long jobs run in the background; webhooks tell you when they finish.
- Voice, music and sound with the audio models.
- 3D assets from text or images with the 3D models.
- Chat and agents with the language models, through the same key.
Every model page lists the model ID, what it takes as input, and its price per run, and the compare pages put two models side by side.
Use FunAI from your AI agent
The FunAI MCP server connects Claude, Cursor, VS Code and other MCP clients to every FunAI model. Connections sign in with OAuth instead of an API key, and each one has its own spending limit.
Security
API keys are shown once when you create them; FunAI stores only a hash and shows the last four characters. Requests go through our API gateway over HTTPS, with rate limits per key. See Security and the Trust Center.