undefined | Better HN

0 pointsmake310mo ago0 comments

I don't think people make the distinction like that. The open source vs non open source distinction boils down to, usually, can you use it for commercial use.

what you're saying is just that it's non reproducible, which is a completely valid but separate issue

0 comments

alpaca12810mo ago

There's already established terms and licenses for non-commercial use. Like "open weights".

Open source has the word "source" in it for a reason, and those models ain't open source and have nothing to do with it.

ben_w10mo ago

Took me until this thread to remember that in the 90s we had "freeware".

piperswe10mo ago

But where's the source? I just see a binary blob, what makes it open source?

jacob01910mo ago

The weights are the source. It isn't as though something was compiled into weights. They're trained directly. But I know what you mean, it would be more open to have the training pipeline and souce dataset available.

timschmidt10mo ago

The weights seem much more like a binary to me, the training pipeline the compiler, and the training dataset the source.

1 more reply

otabdeveloper410mo ago

You can fine-tune their weights and release your own take.

E.g. see all the specialized third-party models out there based on Qwen.

"Open-source" is the wrong word here, what they mean is "you can modify and redistribute these weights".

yetihehe10mo ago

You can also reverse engineer and modify closed source programs (see mods for games). Weights are like compiled version of source data.

2 more replies

microtonal10mo ago

There is work to try to reproduce (the original) R1: https://huggingface.co/open-r1

1una10mo ago

I won't call it "binary blob". Safetensors is just a simple format for storing tensors safely: https://huggingface.co/docs/safetensors/index

j / k navigate · click thread line to collapse