Save 50% or more on cloud cost for Computer Vision
Simplify and automate the deployment of computer vision models like YOLOv8 on 10,000 consumer GPUs on the edge. Get better cost-performance compared to big clouds and high-end GPUs.
Have questions about SCE for your workload?
Book a 15 min call with our team.
Get $50 in testing credits.
Image Tagging
For image tagging applications at scale, Salad’s network of older consumer GPUs offers far higher cost-performance compared to newer, high-end GPUs.
309,000
Images tagged/$
The RTX 2080 on SaladCloud tags 309,000 images per $1, a 60x - 300x cost reduction over managed services like Azure.
30,000
images per dollar
The GTX1650 GPU on SaladCloud processes more images per dollar for ConvNextV2 Tagger V2 compared to any other GPU in the market.
Object Detection
GPUs are faster than CPUs for models like YOLOv8. On Salad, GPUs start at $0.02/hr delivering significant cost-savings and improved performance - the sweet spot between cheaper CPUs and faster high-end GPUs.
73%
Cheaper than Azure
Deploying object detection on SaladCloud is significantly cheaper compared to big box clouds, thanks to the lowest GPU prices in the market.
$0.032
per hour cost
Salad's equivalent to Azure's general compute ($0.19 per hour) for processing a live stream.
Image Tagging: 309,000 Images per $1 - Recognize Anything Model++
Image Segmentation: 50,000 Images per $1 - Segment Anything Model
Analyzing the GTA6 trailer with YOLOv8 on Salad
YOLOv8 Object Detection: 73% cheaper than Azure
Training a custom YOLOv8 model for just $0.25
Image Tagging: Price-performance of 22 GPUs
