canonical: https://jentic.com/apis/offlinecreatorstudio.com/offlinecreatorstudio # OfflineCreator Studio agent commerce The OfflineCreator Studio API runs priced generations against a list of launch models and lets software pay for them per call. A client can lock a quote for a model, aspect ratio, and options, then start the generation against that quote with a maxCredits spend ceiling that refuses an over-budget quote before anything is charged. Without an account, a pay-per-generation route answers with an x402 payment challenge priced from the model list, settles in USDC on retry, and returns a claim token used to poll status and download the output. Credit packs can be bought over x402, and ACP checkout sessions can be created for a generation. ## For AI agents List generation models with their prices, lock a quote, start a generation under a credit ceiling, and poll or download prepaid outputs with a claim token. ## Scope Does not edit uploaded images, host a media library, or manage wallets and payouts. Use for pricing, paying for, and running model generations only. ## Capabilities - List every launch model with its base USD price, credit cost, supported aspect ratios, and payPerGeneration flag - Lock a generation quote for a modelId, aspectRatio, and options before committing credits - Start a generation with a quoteToken and a maxCredits ceiling so an over-budget quote is refused with MAX_CREDITS_EXCEEDED - Pay for a single generation over x402 by retrying with a PAYMENT-SIGNATURE header, with no account needed - Poll a prepaid generation by id with its claim token and fetch the signed output URL once it completes - Buy a credit pack over x402 or open an ACP checkout session for a generation ## Use cases ### Budget-capped generation from an AI agent An agent producing assets for a project needs to generate without overspending. Through Jentic it searches for 'start a priced generation', loads the OfflineCreator Studio startGeneration operation, and first calls quoteGeneration to lock the price for a modelId and aspectRatio. It then starts the generation with the quoteToken and a maxCredits ceiling, so a quote above the ceiling is refused with MAX_CREDITS_EXCEEDED before any credits are reserved. Example prompt: Call POST `/api/v1/generations/quote` with a modelId and aspectRatio '16:9', then POST `/api/v1/generations` with the same modelId, a prompt, the returned quoteToken, and maxCredits 20 ### Model and price discovery Before generating, a tool needs to know which models exist and what each costs. GET `/api/agent/models` returns every launch model with its base price in USD, credit cost, supported aspect ratios, and whether it can be bought anonymously; models that need an uploaded input are marked payPerGeneration false. Because options can change the price, the runtime 402 challenge or a locked quote stays the authoritative amount. Example prompt: Call GET `/api/agent/models` and return the three models with the lowest credit cost that support aspect ratio '1:1' and have payPerGeneration true ### Account-free pay per generation over x402 A client with a USDC wallet can generate without signing up. The first POST `/api/agent/generate` call returns HTTP 402 with x402 payment requirements priced from the model list; retrying with a PAYMENT-SIGNATURE header settles the payment and starts the job. The response carries a claim token, which GET `/api/agent/generate/{id}` accepts as an Authorization Claim header or the claim query parameter to report status and, once done, a short-lived signed output URL. Example prompt: POST `/api/agent/generate` with a modelId and prompt, settle the returned x402 challenge, then poll GET `/api/agent/generate/{id}` with the claim token until it returns the signed output URL ## Key endpoints | Method | Path | Description | | --- | --- | --- | | GET | `/api/agent/models` | List models with prices | | POST | `/api/v1/generations/quote` | Lock a generation quote | | POST | `/api/v1/generations` | Start a priced generation | | POST | `/api/agent/generate` | Pay per generation over x402 | | GET | `/api/agent/generate/{id}` | Poll and download a prepaid generation | | POST | `/api/v1/checkout/x402` | Buy a credit pack over x402 | ## Key resources - **Models** — Public catalog of launch models with base prices, credit costs, aspect ratios, and anonymous-purchase eligibility - **Generations** — Quote locking, credit-backed generation with a spend ceiling, and pay-per-generation jobs polled by claim token - **Checkout** — Credit pack purchase over x402 and ACP checkout sessions for a generation ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 64 / 100 - **Maturity:** AI-Aware - **Dimensions:** - Foundational Compliance: 93 / 100 - Developer Experience & Jentic Compatibility: 62 / 100 - AI-Readiness & Agent Experience: 47 / 100 - Agent Usability: 94 / 100 - Security: 60 / 100 - AI Discoverability: 61 / 100 - **View full report:** https://jentic.com/apis/offlinecreatorstudio.com/offlinecreatorstudio/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring OfflineCreator Studio by hand means sending a Studio bearer key, handling 402 payment challenges, locking quotes before spending, and tracking claim tokens for prepaid jobs. Through Jentic you install once, import OfflineCreator Studio from the API Directory, store the key once, and your agent calls it. - **Permission scoping:** OfflineCreator Studio carries the model and prompt in the request body, so rules bound which operations your agent may call, not which model it uses. You choose the operations, so spending routes like the credit pack purchase and ACP checkout are not included unless you add them. - **Credential handling:** Your Studio key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'start a priced generation' or 'list generation models', and Jentic returns the matching OfflineCreator Studio operation with its input schema, including modelId, quoteToken, and maxCredits, so the agent calls it correctly. ## Related APIs - **Replicate API** — Run open models by version through a prediction API with account billing - **Leonardo.Ai API** — Image generation with model, style, and training controls - **Ideogram.ai API** — Image generation known for rendering text inside images - **Coinbase Advanced Trade** — Crypto trading and account balances on Coinbase ## FAQ ### What authentication does the OfflineCreator Studio API use? Its OpenAPI spec declares HTTP bearer authentication on every operation, carrying a Studio API key or an MCP OAuth access token. Operation descriptions add that the model list and pay-per-generation routes need no account, and that prepaid outputs are fetched with a claim token sent as an Authorization Claim header or the claim query parameter. Through Jentic the Studio key is stored by your own instance and attached at call time. ### Can I cap how many credits a generation spends? Yes. POST `/api/v1/generations` accepts maxCredits, a spend ceiling in credits. If the locked quote is above it, the call is refused with 409 MAX_CREDITS_EXCEEDED before anything is reserved or charged. Lock the price first with POST `/api/v1/generations/quote` and pass the returned quoteToken. ### What are the rate limits for the OfflineCreator Studio API? The OpenAPI spec does not specify rate limits. Check the OfflineCreator Studio documentation at https://offlinecreatorstudio.com/mcp-cli for current limits. ### Is there an OfflineCreator Studio MCP server? You don't need an MCP server to give your agent OfflineCreator Studio. Jentic connects it directly from the API Directory: import it, store your Studio key once, and your agent calls the quote and generation operations on demand without loading another server's tool definitions into its context. ### How do I start a generation through Jentic? Search Jentic for 'start a priced generation', load quoteGeneration to lock a price for your modelId, then execute startGeneration with the prompt, quoteToken, and a maxCredits ceiling. Jentic injects your Studio key on each call. To run it on your own infrastructure, install Jentic One from its GitHub repo. ### How does pay per generation over x402 work? The first POST `/api/agent/generate` call returns HTTP 402 with x402 payment requirements priced from the model list. Retrying with a PAYMENT-SIGNATURE header settles the payment in USDC and starts the generation, with no credit balance or signup involved. The response includes a claim token you present to GET `/api/agent/generate/{id}` to poll status and download the output. ### Can I limit what my agent is allowed to do with the OfflineCreator Studio API? Yes. Write a rule that allows only GET `/api/agent/models`, POST `/api/v1/generations/quote`, and POST `/api/v1/generations`, and the agent can price and run generations against your credits but cannot buy credit packs or open checkout sessions, and every call it makes is logged.