Lumeta for developers

Make images, video and audio from wherever you work

Same Lumeta account, same models, same credits, same Shelf and projects. Just more ways in.

In your AI assistant

Ask Claude, ChatGPT or Cursor for a video in plain words. It finds the photos you saved by name, tells you the price first, and shows the result right in the chat.

Claude, ChatGPT, Cursor, Claude Code, VS Code and more Connect an assistant Pro and Ultimate

In your own app

One simple HTTPS API for every model on Lumeta. See the exact price before you spend, get a webhook when a render is done, and pay with the credits you already have.

REST API, webhooks, OpenAPI Read the API docs Almost here

In your terminal

Local files in, finished files out. Made for scripts, CI and coding agents, with ready-made skills that teach Claude Code, Cursor and Codex to use it well.

The Lumeta CLI and agent skills See the CLI

What every way in shares

One balanceEvery door spends the same credits. If a provider fails, your credits come straight back.
Your Shelf, by nameSay "animate Blue Hat" and the picture you saved as Blue Hat is the one that gets used.
Limits you setSet a credit limit for each key and each connected assistant, and disconnect anything with one click.
All in your libraryEverything you make shows up in your Lumeta library with a small API, MCP or CLI tag.

Try it in a minute

AI assistant: add this as a connector

https://mcp.lumeta.ai

Step by step for each assistant

API: check your balance

curl https://api.lumeta.ai/v1/account \
  -H "Authorization: Bearer lm_live_..."

Make a key under Account → Developer, then make your first image.

Coming from Higgsfield's API? Point your Higgsfield SDK at https://api.lumeta.ai/higgsfield with a Lumeta API key, and the model paths Lumeta also runs keep working as they are. See which ones.

Questions or ideas? Write to [email protected]. How keys, data and limits work: API terms and data.