RunPod Serverless
Pay-per-second worker start to stop: RTX 4000/A4000 16GB $0.00016/s; A100 80GB $0.00076/s (~$2.74/hr); H100 PRO 80GB $0.00116/s (~$4.18/hr); B200 $0.00240/s. Scale-to-zero flex wor
RunPod Serverless scores 4.8/10 for free value in Functions. The main constraint: Paid — From $0.00016/sec (16GB GPU); A100 80GB $0.00076/sec, H100 PRO $0.00116/sec. Pay-per-second worker start to stop: RTX 4000/A4000 16GB $0.00016/s; A100 80GB $0.00076/s (~$2.74/hr); H100 PRO 80GB $0.00116/s (~$4.18/hr); B200 $0.00240/s. Scale-to-zero flex workers.
Free value score
4.8 / 10Show the math
Key metrics
What you get
Paid — From $0.00016/sec (16GB GPU); A100 80GB $0.00076/sec, H100 PRO $0.00116/sec. Pay-per-second worker start to stop: RTX 4000/A4000 16GB $0.00016/s; A100 80GB $0.00076/s (~$2.74/hr); H100 PRO 80GB $0.00116/s (~$4.18/hr); B200 $0.00240/s. Scale-to-zero flex workers.
A100 80GB at ~$2.74/hr flex (active workers ~$2.17/hr with discounts); claims ~25% cheaper than other serverless GPU clouds on flex workers. Excellent
Gotchas
No gotchas documented yet. Contribute one if you know a constraint we missed.
Sources
-
"A100 | 80 GB | $0.00076 | High throughput GPU, yet still very cost-effective. ... H100 PRO | 80 GB | $0.00116"
https://docs.runpod.io/serverless/pricing -
"Save 25% over other Serverless cloud providers on flex workers alone."
https://www.runpod.io/pricing