Inferless
Serverless GPU inference billed per second with scale-to-zero; fractional/shared GPUs available. V2 shared A100 $0.000745/s; also dedicated GPUs (T4, A100, etc.).
Inferless scores 4.8/10 for free value in Functions. The main constraint: Paid — Shared A100 $0.000745/sec (~$2.68/hr). Serverless GPU inference billed per second with scale-to-zero; fractional/shared GPUs available. V2 shared A100 $0.000745/s; also dedicated GPUs (T4, A100, etc.).
Free value score
4.8 / 10Show the math
Key metrics
What you get
Paid — Shared A100 $0.000745/sec (~$2.68/hr). Serverless GPU inference billed per second with scale-to-zero; fractional/shared GPUs available. V2 shared A100 $0.000745/s; also dedicated GPUs (T4, A100, etc.).
Shared/fractional A100 at $0.000745/s plus fast cold starts; fractional GPU support means you only pay for the slice you need - good value for smaller
Gotchas
No gotchas documented yet. Contribute one if you know a constraint we missed.
Sources
-
"Inferless V2:Shared A100 pricing: $0.000745/sec. Note, Inferless also provides fractional GPUs"
https://www.inferless.com/learn/the-state-of-serverless-gpus-part-2 -
"All three top performers (Beam, RunPod, Baseten) cost under $20/month for 100k 1-second inference calls on a T4 GPU."
https://www.beam.cloud/blog/top-serverless-gpu-providers