Replicate, RunPod, and the Commoditization of Inference
You want to run Llama 3. You don't want to set up a server. You don't want to manage GPUs. You don't want to deal with scaling. You want to run it now. You go to Replicate. You paste your prompt. You