Sharing a content piece on,
The State of Serverless GPUs - Part 2
This includes benchmarks around cold-starts, performance consistency, scalability, and cost-effectiveness for models like Llama2 7Bn & Stable Diffusion across different providers.-
inferless.com/learn/the-state-of-serverless-gpus-part-2
Can save months of your evaluation time. Do give it a read!