Show HN: Deepinfra.com Serverless AI model hosting (top models from HF)

[nikola_borisof]

We created a service where you can use the top ML models with a simple API. Models are hosted on our GPU cloud and your can call them via simple HTTP API. This means you can easily build apps with AI, without needing to host any models or running any GPUs. We picked the top 100 models from HuggingFace and made them available on our platform. What other models would you like to see deployed?

[julien_c]

quite cool! haven't tried it yet, but what's the latency on hot-loading a model? (for instance, loading `stabilityai/stable-diffusion-2-1` for the first API call)

[nikola_borisof]

Because this is popular model and many people use it, you will not experience the cold-start latency most likely. But in general it is <10s.