Enterprise Plan
Get Dedicated GPU to run your models with 200% faster speed
What you get
A dedicated GPU and enterprise API, which can generate images in 1.2s, have more concurrency and 100% full privacy.

Models
Load any models from huggingface or from zip file using simple api calls.
- Load model under 3 minutes
- Generate images from any model
- CKPT, Lora, Embeddings, Diffusers, Controlnet Models
- Delete models using API call
- Compiled models for faster generation time
- Model switching in 0.5s

Images
Use img2img, inpainting or text2img using custom trained models with ease.
- 1.2s to generate image
- img2img, text2img or inpainting with any models
- No NSFW filters
- Choose schedulers for each model
- 4K upscaling API
- upto 4 samples at the same time

100% Privacy
Use your own S3 bucket to store images and models
- Use your own S3 bucket
- 100% full privacy
- Store images in your S3 bucket
- Private link protection
- No NSFW filter
- Faster loading of images
Enterprise Pricing
Get inference server with dedicated GPU
Basic Enterprise
For Moderate traffic conditions
- Unlimited Images 😲
- No Rate Limiter 🤩
- 16GB VRAM GPU 😲
- Best for Starters 💻
- Generation time 2s ✈️
- Load 100 Models 🦋
Standard Enterprise
For Startups who want to use ton of models
- Unlimited Images 😲
- No Rate Limiter 🤩
- 24GB VRAM GPU 😲
- High Res Images 💻
- Generation time 1.2s ✈️
- Load 500 Models 🦋
Premium Enterprise
For someone with some serious traffic
- Unlimited Images 😲
- No Rate Limiter 🤩
- 40GB VRAM GPU 😲
- Best for Companies 💻
- Generation time 0.5s ✈️
- Load 1000 Models 🦋
FAQ
The most common questions about Enterprise API and Dedicated GPU
Starts with $499 per month, you can pay yearly and get 20% discount.
No, There is no limit. You can generate as many images as you want.
It takes 1.2s second to generate a image on dedicated GPU. But depends on your image size and steps.
Yes, all images you generate have your copyright. Use it as you like or sell as you like.
After purchase, you will get email with API key and documentation. You can start using API right away.
24X7 support team is available for any issues. Just drop message to support chat on website.
You can upload .ckpt, lora, embeddings, controlnet and diffusers models. You can upload 100+ models.
Yes, there is a queue for API calls. If you make more than 100 API calls per second, it will be queued and processed in order. No API call will be lost.
Want to discuss in person?