EveryFrame API

Use the API console to create a scoped key and manage the separate API balance. The complete endpoint reference and generation examples are at platform.everyframe.studio/docs.

#Choose the right integration

Integration	Authentication	Billing
Developer API	Scoped API key	API balance in USD
Studio MCP or agent CLI	EveryFrame account authorization	Studio credits
everycli miner tooling	Miner hotkey and scoped worker authorization	Your hosting/provider accounts; eligible subnet work

These are separate interfaces. A miner's everycli is not the customer generation API, and API USD funds do not pay for Studio work. The shared account does not merge the balances.

#Read models and quote first

Store the key privately as EVERYFRAME_API_KEY in your server environment. Do not place it in browser code or a distributed mobile app.

bash

curl https://api.everyframe.studio/v1/render/models \
  -H "Authorization: Bearer $EVERYFRAME_API_KEY"

curl https://api.everyframe.studio/v1/render/quote \
  -H "Authorization: Bearer $EVERYFRAME_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"modelId":"seedance-2-0-fast","duration":5,"resolution":"720p"}'

Model discovery and quoting do not create a generation. To render, submit the approved settings and prompt to POST /v1/render, retain the returned ID and poll GET /v1/render/:id until the job finishes. Download the successful output using its returned media URL. A timeout does not prove no request was accepted; inspect the existing job before repeating a paid submission.

#Permissions and spending

Choose the key's allowed operations, expiry and supported monthly spending ceiling in the console. A key acts for its owner. Use the current quote, balance and documented errors rather than converting Studio credits into API dollars. Media links can contain access tokens; keep them private.

For references, supported file upload, errors, limits and exact request/response examples, use the API reference. For account-authorized agents, see Agents & MCP.

#Documentation data

Catalogue tables, plan allowances, top-ups, provider counts and release versions are generated from a reviewed public data snapshot dated 31 October 2026. Builds use that committed snapshot; they do not fetch prices while rendering pages. The snapshot preserves the signed mainnet catalogue and verifies its signature and chain identity.

Snapshots describe the review time. They are not live provider access, capacity, latency or future-price guarantees. Current in-app quotes and everycli offers remain authoritative for new work. Maintainers refresh the snapshot explicitly and compare it with live public sources before deployment; all affected pages and search are rebuilt together.