Which Stable Diffusion version is available?

The text-to-image models run on Krea 2, the 12B open-weight image model — a photoreal merge for general use, an anime style for the anime model and a western-comics finetune for cartoon work — plus Qwen for instruction-based editing. Image-to-image and inpainting on those ids use SDXL companions.

Are prompts filtered?

Prompts reach the model exactly as written. The one automatic block is a local minor-safety keyword check on the prompt, which exists as a legal floor; it makes no vendor calls and applies to nothing else.

Can I use the output commercially?

That depends on the licence of the specific model you called, which the API reports for every model. Check it against your use case — this is your obligation, not something an API can waive for you.

FastGen Cloud vs the Stability AI API

Stability AI publishes a first-party API for its own model family. FastGen Cloud hosts open-weight models — Krea 2 finetunes, Qwen editing, Wan video — behind one contract, and sends your prompts to them unmodified.

Get an API key Quickstart

Side by side

Dimension	FastGen Cloud	the Stability AI API
Model origin	Open-weight community and research models, each with its licence reported by GET /v1/models.	Stability’s own model family, served first-party.
Task coverage	Text-to-image, instruction editing, identity generation, character training and image-to-video in one API.	Image generation and editing across their model line.
Prompt handling	Verbatim — no injected negatives, no added rating or style tags, no rewriting.	Subject to the vendor’s own input policies.
Payment model	Prepaid USD balance, debited per job, auto-refunded on failure.	Credit-based, purchased up front.
Job model	Async everywhere: 202 with a job id, then poll or take a signed webhook.	Mixed synchronous and asynchronous depending on the operation.
Custom subjects	Character training from 4–30 photos, weights held server-side and referenced by id.	Varies by product line.

Comparison of platform design, not of prices — the Stability AI API sets its own rates and changes them independently, so check its current pricing page before deciding. Last reviewed 20 August 2026.

Open weights and licensing

Every model here reports its licence in the API, so you can see exactly what you are calling before you build a product on it. The Qwen editing model is Apache-2.0; the Krea 2 text-to-image finetunes carry the Krea 2 Community License (free commercial use under $1M annual revenue); the identity endpoint depends on a face-embedding component released for non-commercial use.

That last one is a genuine constraint rather than a footnote — check the licence of any model you intend to use commercially, on any platform.

One contract across tasks

Text-to-image, editing, identity, training and video all share the same authentication, the same job envelope, the same idempotency behaviour and the same webhook signature. Adding a second capability to your product is a change of endpoint and body, not a new integration.

What migrating actually involves

The shape is the same everywhere: authenticate, submit a job, wait for a callback, download the output. In practice a migration is one client module and a model-name mapping.

submit + wait

const res = await fetch("https://api.fastgencloud.com/v1/images/generations", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.FASTGEN_API_KEY}`,
    "Content-Type": "application/json",
    "Idempotency-Key": requestId,
  },
  body: JSON.stringify({ model: "general-image", prompt, size: "1024x1024" }),
});
const job = await res.json();          // 202 → { id, status: "queued" }

// then either poll GET /v1/jobs/{id}, or pass webhook_url and get a signed callback

How billing works here

Money is a prepaid balance in your account. You top it up with a card, each job debits it at submit time, and a failed job is refunded automatically — there is no invoice at the end of the month and no way to run up a bill you did not intend.

Rates are per image or per second of video, published on the pricing page. Images above 1 megapixel bill pro rata by pixel count. Outputs are retained for up to 7 days.

Frequently asked questions

Which Stable Diffusion version is available?: The text-to-image models run on Krea 2, the 12B open-weight image model — a photoreal merge for general use, an anime style for the anime model and a western-comics finetune for cartoon work — plus Qwen for instruction-based editing. Image-to-image and inpainting on those ids use SDXL companions.
Are prompts filtered?: Prompts reach the model exactly as written. The one automatic block is a local minor-safety keyword check on the prompt, which exists as a legal floor; it makes no vendor calls and applies to nothing else.
Can I use the output commercially?: That depends on the licence of the specific model you called, which the API reports for every model. Check it against your use case — this is your obligation, not something an API can waive for you.

Start generating

Create an account, add a prepaid balance, and call the API with a key from the console. No subscription, no minimum, no per-seat pricing.

Get an API key