Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Launch HN: Outerport (YC S24) – Instant hot-swapping for AI model weights

news.ycombinator.com

Post reply on HN
Print view View original

11–20 of 25 posts

Re: Launch HN: Outerport (YC S24) – Instant hot-swapping for AI model weights

#11

Post by tovacinni »

CuriouslyC wrote: post #10

This seems useful but honestly I think you guys are better off getting IP protection and licensing out the technology. This is a classic "feature not a product" and I don't see you competing against google/microsoft/huggingface in the model management space.

Maybe! Many people don't want to be vendor locked-in though and there are new GPU cloud providers gaining traction. Some still prefer on-prem.

We hope to make it easier to bridge the multi-cloud landscape by being independent and 'outer'.

Re: Launch HN: Outerport (YC S24) – Instant hot-swapping for AI model weights

#17

Post by tovacinni »

volkopat wrote: post #16

This is really exciting! I was hoping for someone to tackle inference time and this product will definitely be a boost to some of our use cases in medical imaging.

Awesome to hear- that sounds like an application we'd love to help with!

(Please feel free to reach out to us too at towaki@outerport.com !)

Re: Launch HN: Outerport (YC S24) – Instant hot-swapping for AI model weights

#19

Post by phyalow »

Genuine question, whats the difference between your startup and just calling the below code with a different model on a cloud machine, other than some ML/Dev OP's engineer not knowing what they are doing...?

  model = get_model(\*model_config)
  state_dict = torch.load(model_path, weights_only=True)
  new_state_dict = {k.replace('_orig_mod.', ''): v for k, v in state_dict.items()}
  model.load_state_dict(new_state_dict)
  model.eval()
  with torch.no_grad():
  output = model(torch.FloatTensor(X))
  probabilities = torch.softmax(output, dim=X)
  return probabilities.numpy()
Post reply on HN