Generate images with Stable Diffusion and FLUX
Run ComfyUI, Stable Diffusion WebUI Forge or AUTOMATIC1111 in your browser on a real GPU, with nothing to install on your own computer. Generate with SDXL or FLUX, train your own LoRA with Kohya, and download everything you make.
Models download onto the instance at datacenter speed, which is far faster than pulling multi-gigabyte checkpoints over a home connection.
Which GPU to pick
SDXL runs well on 12–16 GB of memory. FLUX and LoRA training are happier on 24 GB, such as an RTX 3090 or RTX 4090. For large batches or video models, choose 48 GB or more.
Generate 1,000 images (Stable Diffusion XL, 1024×1024, 30 steps) costs, on today’s cheapest hosts:
| GPU | Typical time | Cost today |
|---|---|---|
| RTX 3090 | 1.3 h | USD 0.30 |
| RTX 4090 | 0.7 h | USD 0.41 |
| A100 SXM4 | 0.6 h | USD 0.47 |
| L40S | 0.6 h | USD 0.52 |
| RTX A5000 | 1.5 h | USD 0.57 |
Ready-made software
Pick one of these when you rent and it starts with everything installed:
- ComfyUI (image & video generation) — Node-based Stable Diffusion / FLUX / video workflows. Needs a CUDA 12.1+ host.
- Stable Diffusion WebUI Forge — Forge WebUI for Stable Diffusion and FLUX image generation.
- Stable Diffusion WebUI (A1111) — The classic AUTOMATIC1111 WebUI for Stable Diffusion.
- Kohya's GUI (LoRA / model training) — Train LoRAs and fine-tune Stable Diffusion / FLUX with the Kohya_ss GUI. TensorBoard included.
How it works
- Create an account with Google or your email address. There is no ID upload: your first payment is how we know who you are.
- Add credits in your own currency.
- Add your SSH key, then pick the software and a GPU from the live board.
- Wait for Running — about a minute on a warm host, a few minutes on a cold one — then open the app or connect over SSH.
- Stop when you are done. GPU billing ends at once; only the disk is billed until you destroy the instance.
Questions
Do I need to install anything?
No. Pick ComfyUI, Forge or A1111 as the software when you rent, then open the app link from the instance card in your browser.
Can I use my own models and LoRAs?
Yes. Upload them, or download them on the instance directly from Hugging Face or Civitai.
Who owns the images I generate?
You do. We do not look at or keep what you generate. Your use must follow the acceptable-use policy.