H100 Replicate VS RunPod

We provide a go_fast flag within the API which toggles a version of flux-schnell optimized for inference. Currently this version is a compiled fp8 quantization with an optimized attention kernel. We’ll update the model and this documentation as we develop further enhancements.

We provide a go_fast flag within the API which toggles a version of flux-schnell optimized for inference. Currently this version is a compiled fp8 quantization with an optimized attention kernel. We’ll update the model and this documentation as we develop further enhancements.

Fast booting fine-tunes

Sometimes, we’re able to optimize how a trained model is run so it boots fast. This works by using a common, shared pool of hardware running a base model. In these cases, we only ever charge you for the time the model is active and processing your requests, regardless of whether or not it’s public or private.

Fast booting fine-tunes are labeled as such in the model’s version list. You can also see which versions support the creation of fast booting models when training.

Fast booting fine-tunes

Sometimes, we’re able to optimize how a trained model is run so it boots fast. This works by using a common, shared pool of hardware running a base model. In these cases, we only ever charge you for the time the model is active and processing your requests, regardless of whether or not it’s public or private.

Fast booting fine-tunes are labeled as such in the model’s version list. You can also see which versions support the creation of fast booting models when training.