01Video · 1.0
Centaurus 1.0
Cinematic text-to-video and image-to-video. 720p / 1080p, 2–15s clips.
Text to VideoImage to Video
from $0.20per second
Docs →Generate anything.
No filters. No limits.
The unfiltered image & video generation API.
One request, no polling. Pay per call.
Generations today14,283,129
01Models3 families
State-of-the-art models, unrestricted.
Three families across video, image and editing. One API, pay only for what you generate.
02Image + Video · 1.0
Hera 1.0
Text-to-image, prompt-driven image editing, and multi-shot image-to-video.
Text to ImageImage EditImage to Video
from $0.056per image
Docs →03Video · 1.0
Zeus 1.0
High-quality image-to-video with smooth motion. 480p / 720p, 5s or 8s.
Image to VideoImage to Video 480p
from $0.30per run
Docs →02Why Api1804 reasons
Built for builders, not policy teams.
Plain HTTP. Real infra. No moderation surprises.
01
Zero refusals
No safety filters. No policy 400s. No silent prompt rewrites.
02
Sub-3s latency
H100 fleet, four regions. p95 under 2.8s for 1024px.
03
One call, no SDK
POST to /v1/generate?wait=true and get the finished URL back. Plain HTTP.
04
Pay per call
No subscription. No minimums. Failed generations are auto-refunded.
03IntegrationOne request
One call. No polling.
A Bearer key and a JSON POST. Use ?wait=true and the finished URL comes right back.
01
Get your API key
Sign up and copy your key from the dashboard. New accounts get $10 to test.
02
POST your prompt
Send one request to /v1/generate with a Bearer key. Add ?wait=true to get the result in a single call.
03
Read the docs →Get the URL
Read the output from data[0].url. No SDK, no polling loop required.
node.js · v20
// One request. The finished URL comes back. const res = await fetch("https://api18.dev/v1/generate?wait=true", { method: "POST", headers: { Authorization: `Bearer ${process.env.API18_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ model: "centaurus-1.0-text-to-video", prompt: "a moody noir street, neon rain", resolution: "720p", }), }); const job = await res.json(); console.log(job.data[0].url); // → https://…/your-video.mp4
POST/v1/generateAny model
POST/v1/video/generationsVideo
POST/v1/images/generationsImages
GET/v1/jobs/:idPoll status
04Pricing
Pay only for what you generate.
No subscription. No minimums. Failed requests are free.
Text → Videocentaurus-1.0-text-to-video
720p · 2–15s clips
$0.20/ second
Image → Videocentaurus-1.0-image-to-video
720p · 3–15s clips
$0.28/ second
Image → Videohera-1.0-image-to-video
720p · 5/10/15s clips
$0.20/ second
Image Edithera-1.0-image-edit
1–3 reference images · prompt-driven
$0.07/ run
Text → Imagehera-1.0-text-to-image
1k/2k · 9 aspect ratios
$0.056/ image
Image → Videozeus-1.0-image-to-video
480p · 5s clip
$0.30/ run
Image → Videozeus-1.0-image-to-video-480p
480p · 5s clip
$0.40/ run
What is Api18.dev?+
Api18.dev is an uncensored REST API for AI image and video generation. It applies no content moderation to prompts and supports adult/+18 and NSFW use cases. You send one JSON POST with a Bearer key and receive a media URL — text-to-image, image editing, text-to-video and image-to-video.
Is the API really uncensored?+
Yes. There is no prompt-moderation classifier and adult/+18 use is permitted in the terms. Your prompt is passed to the model and you get the output. Illegal content remains prohibited, and you are responsible for lawful use.
How much does it cost and is there a free trial?+
Pricing is pay-as-you-go in USD per call, with no subscription or minimums, and failed calls are refunded automatically. New accounts get $10 in free credit to test every model in the dashboard playground. The API runs on paid balance.
How do I pay for the API?+
You top up your balance with USDT on the Solana or Tron network — no card and no identity verification required. Top-ups confirm in seconds, and because USDT is dollar-pegged it maps directly to each call's USD price.
Do I need an SDK or a polling loop?+
No. Api18.dev is a plain REST API. The simplest integration is one call to POST /v1/generate?wait=true, which returns the finished job with the output at data[0].url. Any HTTP client works, and an LLM-ready reference lets ChatGPT, Claude, Grok or Gemini write the integration for you.
What can it generate?+
Text-to-image, prompt-driven image editing, text-to-video, and image-to-video (animating a still you provide). Each model's exact parameters, pricing and a runnable example are on the models page.
Build the thing
others won't host.
Free credits on signup. Production-ready in sixty seconds.