Omg, it’s almost like AI models are already commodities.
davehorse on
Please be good for coding
crazybmanp on
This isn’t really open. It’s non-commercial so you would need to go buy a card to run this on because no one can sell you it and the cards are expensive
johnryan433 on
This is so bullish nvidia releases more open source models that just require more vram in turn requiring more GPUs from Nvidia , that’s a 4d chess move right there. 🤣😂
KrypXern on
Isn’t this just a Qwen 2 finetune? Hardly meets the hype the article is trying to sell
ChocolateBunny on
I don’t feel like this is a big deal. It seems like they compared it to llama 3.1 405B which is also “open source”. It seems like Nvidia published the weights and promises to publish the training algorithm. I believe nVidia is currently under a lawsuit for using copyrighted training data so I would be careful with whatever you use this stuff for.
Enjoy-the-sauce on
I can’t wait to see which one destroys civilization first!
starttupsteve on
Why is nvidia dumping so much money into OpenAI then?
imaginary_num6er on
*”Our logic is: If we don’t build it, they can’t come.”* – Jensen Huang
guitarokx on
What I don’t understand is why they are a major investor in OpenAI then?
chameleon_circuit on
Odd because they just invested in OpenAi during the most recent round of funding.
XXsforEyes on
Zoiks! Gonna have to look into this more!!
Jdunc97 on
I was too busy finishing middle school to buy nvidia stock…. Too late to hop on?
16 Comments
Oof. Nvidia is known for anticompetitive behavior.
Them controlling the hardware could be dangerous for the industry
Absolutely classic “commoditize your complement” play. [https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/](https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/)
Omg, it’s almost like AI models are already commodities.
Please be good for coding
This isn’t really open. It’s non-commercial so you would need to go buy a card to run this on because no one can sell you it and the cards are expensive
This is so bullish nvidia releases more open source models that just require more vram in turn requiring more GPUs from Nvidia , that’s a 4d chess move right there. 🤣😂
Isn’t this just a Qwen 2 finetune? Hardly meets the hype the article is trying to sell
I don’t feel like this is a big deal. It seems like they compared it to llama 3.1 405B which is also “open source”. It seems like Nvidia published the weights and promises to publish the training algorithm. I believe nVidia is currently under a lawsuit for using copyrighted training data so I would be careful with whatever you use this stuff for.
I can’t wait to see which one destroys civilization first!
Why is nvidia dumping so much money into OpenAI then?
*”Our logic is: If we don’t build it, they can’t come.”* – Jensen Huang
What I don’t understand is why they are a major investor in OpenAI then?
Odd because they just invested in OpenAi during the most recent round of funding.
Zoiks! Gonna have to look into this more!!
I was too busy finishing middle school to buy nvidia stock…. Too late to hop on?
soooooo…its crap?