4 ms·

Nvidia has hardly open sourced the nvcc compiler, yet no-one here is complaining about why it is closed source. Modular on the other hand creates the Mojo comp

by rvz 2mo ago

Nvidia has hardly open sourced the nvcc compiler, yet no-one here is complaining about why it is closed source. Modular on the other hand creates the Mojo compiler gets criticised for not open sourcing it immediately and now once they do, no-one cares anymore. Huggingface was not just a target for open source, but as a force to have open weight models run better on Nvidia against the rest.

paufernandez 2mo ago
I'm still rooting for Modular, though.
silisili 2mo ago
I assume it's because nvidia is a hardware company that produces world class hardware, and it's a compiler to target that. Whereas mojo is a general purpose language, and we're absolutely spoiled for choice on modern languages with open source compilers. I'm not saying it's fair or right, I still think mojo is neat, but isn't exactly comparing apples to apples.
- rvz 2mo ago
  We can stop with these weak excuses since AMD and Intel have done more for open source than Nvidia has, including their GPU drivers for Linux. Nvidia on the other hand has not and the best they have done is a bunch of closed-source blobs which they do more closed source releases than the rest. Mojo is open source and targets all GPU architectures for their compiler regardless of the vendor and nvcc targets their own (and both that and CUDA are closed source). So this is directly an apples to apples comparison.
  - silisili 2mo ago
    I'll give credit to Intel, they've always been open source friendly and I prefer to buy them because I'm not a gamer, so don't need the best. AMD is another story. Their binary closed source blobs were even worse than nvidias for many years. So much so people used terrible performance open source tries just to avoid the headache. That they finally slopped together an open source version after they'd lost the GPU race isn't exactly noble. But as an open source supporter in general, I commend the effort still.
latchkey 2mo ago
> no-one cares anymore. possibly because it took qualcomm buying them to make that happen.
- rvz 2mo ago
  Just say you don't know. Mojo was partially open source before Qualcomm bought them, and they were going to do open source it anyway. Was NVCC or CUDA ever open source since the lifetime of its development?
  - latchkey 2mo ago
    > they were going to do open source it anyway. uh huh. > Was NVCC or CUDA ever open source since the lifetime of its development? you ever ask Nvidia why? i did.
    - rvz 2mo ago
      > uh huh. Exactly. You're one of those that don't care. Mojo's standard library was open sourced a year before the acquisition. > you ever ask Nvidia why? i did. The correct answer to my question is "No". CUDA or NVCC was never open sourced during its lifetime. Since you "claim" to have asked them and know the true reason why is not open source, just say what Nvidia told you verbatim right here. Intel and AMD have no problem open sourcing their GPU drivers for Linux, but Nvidia has a problem with that. We can only speculate that "open sourcing" it reveals their GPU trade secrets and their intellectual property for others to copy.
      - latchkey 2mo ago
        I care, I just don't believe you in that statement. There were "confidential nda" notes embedded in the release that Chris didn't remove until afterwards. That doesn't scream: "we're going to open source this some day!" If it was going to "eventually" be made open source, why not do it from the start? Why not do it a year in? From what I was told, your speculation is incorrect. I can't say anything more than that, sorry. I don't know why you're so aggressive in tone towards me, chill out.
chii 2mo ago
> as a force to have open weight models run better on Nvidia against the rest. this is the crux - if nvidia makes it so that open weights end up running better on nvidia hardware than competitor's, then it's going to prevent hardware innovation and competitiveness in the entire sector. It's like as tho General Motors buys out oil refinery to make gas for all, but the gas somehow runs smoother in GM cars.
- coderenegade 2mo ago
  It's a clever move if that's what they're doing. They're restricted in China, and are likely to face stiff competition from Chinese chipmakers in the coming years. Acquiring the largest repository of trainable models and ensuring they run better on Nvidia hardware is probably one of the few moves they have for keeping ahead of the competition. I mean, it would be terrible for the consumer, but it does make me think that NVDA is a decent investment.
  - high_na_euv 2mo ago
    Why would they have to acquire hf in order to optimize their stack for those models? Couldnt they just take those models anyway?
    - coderenegade 2mo ago
      Well now that they own it, they can do whatever they like. They don't necessarily have to be optimizing the models for their hardware, they could just setup preferential pipelines to funnel users to their ecosystem. That seems fairly prudent given the challenges coming their way.
- robocat 2mo ago
  If you don't own the customer relationship, you don't own the profit. See App store, FB. NVidia is delivering commodity hardware to the hyperscalers, and would eventually get commodity margins (when hyperscalers make their models work on their own hardware). Amazing article on relevant economics of squeezing vendors - actually about antitrust ad-models but: The answer is that [advertising based] companies are an ideal test case for an increasingly common business meta-model, where companies try to create a consumer surplus at one end in order to maximize their negotiating leverage for capturing the producer surplus everywhere else in their supply chain. https://www.thediff.co/archive/ad-supported-platforms-are-a-unique-antitrust-problem https://www.thediff.co/archive/ad-supported-platforms-are-a-... Perhaps not relevant to huggingface, sorry.

by rvz 2mo ago

paufernandez 2mo ago

I'm still rooting for Modular, though.

silisili 2mo ago

I assume it's because nvidia is a hardware company that produces world class hardware, and it's a compiler to target that. Whereas mojo is a general purpose language, and we're absolutely spoiled for choice on modern languages with open source compilers. I'm not saying it's fair or right, I still think mojo is neat, but isn't exactly comparing apples to apples.

rvz 2mo ago
We can stop with these weak excuses since AMD and Intel have done more for open source than Nvidia has, including their GPU drivers for Linux. Nvidia on the other hand has not and the best they have done is a bunch of closed-source blobs which they do more closed source releases than the rest. Mojo is open source and targets all GPU architectures for their compiler regardless of the vendor and nvcc targets their own (and both that and CUDA are closed source). So this is directly an apples to apples comparison.
- silisili 2mo ago
  I'll give credit to Intel, they've always been open source friendly and I prefer to buy them because I'm not a gamer, so don't need the best. AMD is another story. Their binary closed source blobs were even worse than nvidias for many years. So much so people used terrible performance open source tries just to avoid the headache. That they finally slopped together an open source version after they'd lost the GPU race isn't exactly noble. But as an open source supporter in general, I commend the effort still.

latchkey 2mo ago

> no-one cares anymore. possibly because it took qualcomm buying them to make that happen.

rvz 2mo ago
Just say you don't know. Mojo was partially open source before Qualcomm bought them, and they were going to do open source it anyway. Was NVCC or CUDA ever open source since the lifetime of its development?
- latchkey 2mo ago
  > they were going to do open source it anyway. uh huh. > Was NVCC or CUDA ever open source since the lifetime of its development? you ever ask Nvidia why? i did.
  - rvz 2mo ago
    > uh huh. Exactly. You're one of those that don't care. Mojo's standard library was open sourced a year before the acquisition. > you ever ask Nvidia why? i did. The correct answer to my question is "No". CUDA or NVCC was never open sourced during its lifetime. Since you "claim" to have asked them and know the true reason why is not open source, just say what Nvidia told you verbatim right here. Intel and AMD have no problem open sourcing their GPU drivers for Linux, but Nvidia has a problem with that. We can only speculate that "open sourcing" it reveals their GPU trade secrets and their intellectual property for others to copy.
    - latchkey 2mo ago
      I care, I just don't believe you in that statement. There were "confidential nda" notes embedded in the release that Chris didn't remove until afterwards. That doesn't scream: "we're going to open source this some day!" If it was going to "eventually" be made open source, why not do it from the start? Why not do it a year in? From what I was told, your speculation is incorrect. I can't say anything more than that, sorry. I don't know why you're so aggressive in tone towards me, chill out.

chii 2mo ago

> as a force to have open weight models run better on Nvidia against the rest. this is the crux - if nvidia makes it so that open weights end up running better on nvidia hardware than competitor's, then it's going to prevent hardware innovation and competitiveness in the entire sector. It's like as tho General Motors buys out oil refinery to make gas for all, but the gas somehow runs smoother in GM cars.

coderenegade 2mo ago
It's a clever move if that's what they're doing. They're restricted in China, and are likely to face stiff competition from Chinese chipmakers in the coming years. Acquiring the largest repository of trainable models and ensuring they run better on Nvidia hardware is probably one of the few moves they have for keeping ahead of the competition. I mean, it would be terrible for the consumer, but it does make me think that NVDA is a decent investment.
- high_na_euv 2mo ago
  Why would they have to acquire hf in order to optimize their stack for those models? Couldnt they just take those models anyway?
  - coderenegade 2mo ago
    Well now that they own it, they can do whatever they like. They don't necessarily have to be optimizing the models for their hardware, they could just setup preferential pipelines to funnel users to their ecosystem. That seems fairly prudent given the challenges coming their way.
robocat 2mo ago
If you don't own the customer relationship, you don't own the profit. See App store, FB. NVidia is delivering commodity hardware to the hyperscalers, and would eventually get commodity margins (when hyperscalers make their models work on their own hardware). Amazing article on relevant economics of squeezing vendors - actually about antitrust ad-models but: The answer is that [advertising based] companies are an ideal test case for an increasingly common business meta-model, where companies try to create a consumer surplus at one end in order to maximize their negotiating leverage for capturing the producer surplus everywhere else in their supply chain. https://www.thediff.co/archive/ad-supported-platforms-are-a-unique-antitrust-problem https://www.thediff.co/archive/ad-supported-platforms-are-a-... Perhaps not relevant to huggingface, sorry.