1] Hyperbolic
Focus: AI inference & model rental
Compute Model: Centralized GPU marketplace
Specialization/Edge: Low-cost GPU rental, Hugging Face integration
Integration Method: REST/API
Website: app.hyperbolic.ai
2] Cherry Servers
Focus: AI/ML training
Compute Model: Bare-metal GPU servers
Specialization/Edge: Fast deploy, customizable Nvidia/AMD hardware
Integration Method: Portal & REST API
Website: cherryservers.com
3] SurferCloud
Focus: AI inference & GPU workloads
Compute Model: GPU UHost (vGPU cloud)
Specialization/Edge: Affordable RTX 4090/Tesla P40, elastic vGPU scaling
Integration Method: Portal/API
Website: surfercloud.com
4] Novita AI
Focus: LLM inference & API
Compute Model: Serverless GPU + VM instances
Specialization/Edge: Spot pricing (~50%) on GPUs, LLM-serving endpoints
Integration Method: API/SDK + hosted UI
Website: novita.ai
5] Vultr
Focus: AI training & inference
Compute Model: Virtual GPU VMs (A100, AMD MI)
Specialization/Edge: Scalable GPU tiers, Kubernetes-ready
Integration Method: Console/CLI/API
Website: vultr.com
6] Immers.cloud
Focus: AI training & inference
Compute Model: Bare-metal GPU VMs
Specialization/Edge: Dedicated Tesla/A100/H100, per-second billing, no oversubscription
Integration Method: Portal/API (OpenStack Yoga)
Website: immers.cloud
7] Database Mart
Focus: AI training & inference
Compute Model: Bare-metal GPU VMs
Specialization/Edge: Wide GPU config (H100/A100/RTX), cost-effective, rapid deployment
Integration Method: Portal, control panel
Website: databasemart.com
8] Vast.ai
Focus: AI model training, LLM inference, 3D rendering
Compute Model: Virtualized GPU instances (serverless-style, pay-as-you-go)
Specialization/Edge: Spot pricing for GPUs, flexible scheduling, lifetime referral rewards
Integration Method: Web dashboard + API/CLI
Website: vast.ai
9] RunPod
Focus: AI/ML workloads, LLM training, batch rendering
Compute Model: GPU pods + serverless GPU instances
Specialization/Edge: Low-latency endpoints, containerized workloads, referral + affiliate program
Integration Method: Web dashboard + API/SDK + container support
Website: runpod.io
More compute providers can be found in this Google Sheet.
Do let us know in the comments area below if we missed something! :)
