Live data from Hacker News

Meta to release open-source commercial AI model

zdnet.com

31–40 of 168 posts

Re: Meta to release open-source commercial AI model

#31

What's the monetization model here? Is this a closed-source version of their open-source model? (That's suggested by the phrase in the article, "a commercial version of LLaMA, its open-source large language model".)

Maybe they just want the de facto standard LLM to be one that only says nice things about facebook and Zuckerberg?

Re: Meta to release open-source commercial AI model

#34
post #27

Maybe they've solved the fingerprinting problem and can identify text generated from their model, and this is a way of discovering the market they can sell more advanced models to directly. B2B leadgen...

I mean you could probably just train it on some sequence s.t. the model identifies itself, would be hard to detect that

Re: Meta to release open-source commercial AI model

#35

What's the monetization model here? Is this a closed-source version of their open-source model? (That's suggested by the phrase in the article, "a commercial version of LLaMA, its open-source large language model".)

Meta is a company that makes money off of users endlessly browsing content. It would follow that making it easier/faster to generate content would benefit Meta.

Re: Meta to release open-source commercial AI model

#36
post #16

It's not open source, it's freeware or something like that. Weights aren't the source code of LLMs, they're the binaries.

Thank you for succinctly explaining the difference, I learned something today

Compiling source code doesn't cost million of dollars though

Re: Meta to release open-source commercial AI model

#37
post #27

Maybe they've solved the fingerprinting problem and can identify text generated from their model, and this is a way of discovering the market they can sell more advanced models to directly. B2B leadgen...

I don't think so because I believe you can train AI models against other AI models. I believe you can fingerprint a family of models, but that's not going to tell whether you just used the general approach outlined in the academic papers.

Re: Meta to release open-source commercial AI model

#38

Earlier quoted context omitted.

Thank you for succinctly explaining the difference, I learned something today

Compiling source code doesn't cost million of dollars though

Yea the weights are the secret sauce that OpenAI and competitors generally protect.

Re: Meta to release open-source commercial AI model

#39

Earlier quoted context omitted.

Thank you for succinctly explaining the difference, I learned something today

Compiling source code doesn't cost million of dollars though

That doesn't change the meaning of Open Source. These are "free as in beer", not "free as in [modify the sources and rebuild it]". There are LLMs for which that is true, which include a specific list of training data. If you wanted to "uncensor" one of those, you could curate the source data and rebuild it, instead of trying to get it to unlearn what it was taught.

Re: Meta to release open-source commercial AI model

#40

What's the monetization model here? Is this a closed-source version of their open-source model? (That's suggested by the phrase in the article, "a commercial version of LLaMA, its open-source large language model".)

Based on the podcast with Lex Friedman and Mark Zuckerberg, see ~minute 30. My hypothesis based on the context of Mark discussing the release is that it's going to be completely open source and can licensed to be used commercially. Not that Meta is going to add a whole new revenue side of business to compete with OpenAI. i.e. "Here is model, with commercially permissive licensing" not "Here is model that you can use…

Another hypothesis is that they are trying to rehab their brand.

They can even write it as 'good will' on their financial statements.

It kind of is working.

Post reply on HN