Salad

Provides a distributed GPU cloud that supplies low-cost compute for AI/ML inference, batch processing, image generation, and other GPU-intensive workloads to address high costs and limited supply from hyperscalers. Serves developers, data teams, and operations/IT teams in businesses that need scalable production AI deployment. Positioned as a lower-cost alternative to hyperscalers and datacenter GPUs by aggregating 60,000+ daily active consumer GPUs across 191 countries starting at $0.02/hour, claiming up to 90% cost savings.

Key features

• Fully managed scalable container engine
• Flexible cost-effective GPU community cloud
• Enterprise-grade datacenter secure cloud
• Dedicated proxies in ~200 countries
• Speech-to-text transcription API
• K8 pods as container deployments via virtual kubelets
• Distributed file storage
• Object storage
• Managed databases