GPU marketplace
RunPod
Rent selectable GPUs through Pods, or run autoscaling inference workers through the separate Serverless product.
Product facts checked October 2, 2026Editorial scores last reviewed September 3, 2026
Some outbound links use a first-party redirect hop for click counting. Commission is only claimed when a partner programme is active for that specific link — most vendor hops here are not paid placements. Affiliate disclosure.
Marketplace referrals (for example RunPod or Vast) may return credits or kickbacks when a programme is active — that is a material connection even when it is not a cash CPA.
Best for
Teams that want to pick a specific GPU SKU and control the container stack
Watch out
Pod compute stops when a Pod stops, but persistent volume disks ($0.20/GB-month stopped) and network volumes ($0.07/GB-month under 1TB; $0.05/GB-month above 1TB) continue billing until deleted. Temporary container disk is erased on stop. Serverless worker billing and always-on worker settings differ from Pod billing.
| Billing | Pods: per-second compute and local disk billing; network volumes: hourly storage billing. Serverless Flex/Active workers: per-second lifecycle time. |
|---|---|
| GPU choice | Broad consumer and datacenter SKUs |
| Cold start | Depends on pod spin-up and image cache |
| Model access | Bring your own weights or pull from registries |
Sources for the dated product and announcement claims
- runpod official product and billing documentation — source checked October 2, 2026
- runpod official product and billing documentation — source checked October 2, 2026
- runpod official product and billing documentation — source checked October 2, 2026
- runpod official product and billing documentation — source checked October 2, 2026
Fit detail
When RunPod is the right shortlist — and when it is not
Use these lists to kill bad comparisons early, before you compare logos.
Ideal for
- Picking a specific consumer or datacenter GPU SKU
- Self-hosted vLLM / TGI / Comfy stacks you already trust
- Burst experiments that still need SSH-level control
Usually not ideal for
- Shipping a chat API tomorrow with no container skills
- Teams that cannot monitor idle pods overnight
- Strict multi-region SLAs without your own ops layer
Spend & ops
How cost behaves — and what breaks first
No invented $/hour quotes. These notes explain the failure modes that show up on the first real invoice.
Cost mental model
Pods bill allocated hardware time, including idle time, with storage charged separately. Serverless bills worker lifetime rounded up to a second, including startup, processing and idle timeout (5 seconds by default). Flex scaling and Active workers have different utilization costs.
Ops notes
- Stop Pods to end compute; delete unneeded retained volume disks and network volumes to end storage charges.
- Serverless Flex workers scale to zero; Active workers stay running. Worker startup/model loading, processing and the configured idle window are billable.
- A zero credit balance stops network-volume-backed Pods but terminates Pods without network volumes; data on terminated Pods is unrecoverable. Monitor the balance before relying on retained data.
Fit scores
RunPod on the decision axes
Same editorial scale as the hub chart — useful for shortlists, not as a price quote.
RunPod
- GPU choice9/10
- Time-to-serving (editorial)6/10
- Price clarity8/10
- Production ops6/10
- Open-model breadth8/10
Among all 16 providers
GPU choice
Pick exact GPUs
Time-to-serving (editorial)
Warm-path fit
Price clarity
Easy to forecast
Production ops
Less DIY ops
Open-model breadth
Catalog depth
Scores are editorial planning ratings (1–10) for product shape — not published $/hour quotes or vendor SLAs. Verify current pricing on each provider site.
Compare
RunPod vs alternatives
Open a head-to-head when you are deciding between product shapes.
Related
Keep the decision attached to the rest of the stack
Cloud rental is one path. Local VRAM and open-weight fit still matter.