The Samsa API lets you use Samsa outside the app — generate images and videos, train and manage your custom models, and check your credit usage from your own tools, scripts, or automations. You can also connect AI assistants like Claude and ChatGPT to Samsa over MCP, so they can create content for you directly in a chat.
This collection walks you through the parts you manage inside Samsa. For the full technical reference — endpoints, request formats, and code examples — head to the Samsa developer docs (opens in a new tab).
Generate images from a prompt, optionally using your trained models
Edit images with Magic Edit
Generate videos from a start frame, from text, or from text styled with your models
Train and manage models — create, list, update, and delete your custom models
Check usage — read your organization's credit balance and usage
Organization owners and admins create and manage the API keys that unlock the API.
Anyone you share a key with (or any AI assistant you connect) then works within that key's permissions and your organization's credits.
API access is tied to your organization, not a personal account — keys, credits, and usage all belong to the organization.
The API is job-based: when you ask it to generate an image or video, it accepts the request and hands back a job you can check on until it's ready. Everything you create costs the same credits as it does in the app, drawn from your organization's pool (and any team budgets you've set up).
Direct API — use an API key from any tool or script that can make web requests. Start with Creating and managing API keys.
MCP — connect an AI assistant (Claude, ChatGPT, and more) so it can use Samsa on your behalf. See Connecting an MCP client.
Set up access: Creating and managing API keys
Connect an assistant: Connecting an MCP client (Claude, ChatGPT and more)
Understand costs: API usage, credits and rate limits
Get notified when jobs finish: Webhooks
Full developer reference: the Samsa developer docs, including the Quickstart and API reference
