Popular models, deployed in a few clicks.
Recipes are pre-configured container groups for models people actually run. Sign in, pick a recipe, choose GPUs and replica count, customize if you want, and start serving on the lowest-priced GPUs in the market.
Start from a recipe.
Hourly prices reflect the recipe's default GPU at Lowest priority; change either in the portal.
FLUX.1-Schnell on ComfyUI
Fast, high-quality image generation with a stateless ComfyUI API. 5,243 images per dollar in our benchmark.
Stable Diffusion XL with Refiner
SDXL base plus refiner on ComfyUI. 3,405 images per dollar in our benchmark.
Dreamshaper 8 on ComfyUI
A popular SD 1.5 fine-tune for stylized generation at very low cost.
YOLOv8-M
Real-time object detection; 73% cheaper than Azure for live streams in our benchmark.
Ollama · Llama 3.1
An OpenAI-compatible endpoint for Llama 3.1 on a flat hourly rate. The default for OpenClaw and Cline setups.
TGI · Mistral 7B Instruct
Text Generation Inference serving Mistral 7B for chat and agents.
Don't see your model? Bring any Docker image to Salad Container Engine, or call open models per token through AI Gateway.
Stop overpaying for cloud.
"By switching to SaladCloud, Civitai is now serving inference on over 600 consumer GPUs to deliver 10 Million images per day and training more than 15,000 LoRAs per month."
Justin Maier · Founder, Civitai"On SaladCloud's consumer GPUs, we are running 3X more scale at half the cost of A100s on our local provider and almost 85% less cost than the two major hyperscalers we were using before."
Jamsheed Kamardeen · CTO, Blend"Salad makes it more realistic to keep up with deploying these new models. We might never deploy most of them if we had to pay AWS cost for them."
Shawn Rushefsky · Founder, Dreamup