top 43 comments

sorted by: hot top controversial new old
[–] 14 points 1 year ago* (last edited 1 year ago) (5 children)

Where can I buy this?

Edit: I realized after I commented this was the product page.. My bad. It was more of a take my money now scenario

  • source
  • hideshow 5 child comments
  • [–] 6 points 1 year ago (2 children)

    i wonder if the driver to run is compatible with linux.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 6 points 1 year ago

    These only work with ARM cpus I think

  • source
  • [–] 2 points 1 year ago (1 child)

    Does anyone know if it can run CUDA code? Because that's the silver bullet ensuring Nvidia dominance in the planet-wrecking servers

  • source
  • hideshow 1 child comment
  • [–] 5 points 1 year ago

    llama and pytorch support it right now. CUDA isn't available on its own as far as I can tell. I'd like to try one out but the bandwidth seems to be ass. About 25% as fast as a 3090. It's a really good start for them though.

  • source
  • parent
  • [–] 6 points 1 year ago

    I kinda want an individual consumer-friendly, low-end/mid-end alternative that can run my games and video editing software for very small projects.. so far I'm only eyeing the Lisuan G100, which seems to fit that bill..

    This seems cool though, other than AI, it could be used for distributed cloud computing or something of that sort

  • source
  • [–] 6 points 1 year ago (9 children)

    For inference only. NVIDIA GPU's are so big because they can train models. Not just run them. All other GPU's seem to lack that capacity.

  • source
  • hideshow 9 child comments
  • [–] 5 points 1 year ago (8 children)

    You can train or fine-tune a model on any GPU. Surely, It will be slower, but higher VRAM is better.

  • source
  • parent
  • hideshow 8 child comments
  • [–] 0 points 1 year ago (7 children)

    No. The CUDA training stuff is Nvidia only.

  • source
  • parent
  • hideshow 7 child comments
  • [–] 3 points 1 year ago (6 children)
  • [–] 0 points 1 year ago* (5 children)

    AMD has been lying about that every year since 2019.

    Last time I checked it didn't. And it probably still doesn't.

    People aren't buying NVIDIA if AMD would work too. The VRAM prices NVIDIA asks are outrageous.

  • source
  • parent
  • hideshow 5 child comments
  • [–] 2 points 1 year ago (4 children)

    I run llama.cpp and PyTorch on MI300s. It works really well.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 1 point 1 year ago* (3 children)

    Can you train on it too? I tried Pytorch on AMD once and it was awful. They promised mountains but delivered nothing. Newer activation functions were all broken.

    llama.cpp is inference only, for which AMD works great too after converting to ONNX. But training was awful on AMD in the past.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 1 year ago (2 children)

    We have trained transformers and diffusion models on AMD MI300s, yes.

  • source
  • parent
  • hideshow 2 child comments
  • [+] -6 points 1 year ago (5 children)

    PCI-E 3.0, DDR4 memory, no drivers, no fans You would be better off any DDR4 CPU with a bunch of ram

  • source
  • hideshow 5 child comments
  • [–] [S] 14 points 1 year ago (4 children)

    When you definitely know the difference between what a CPU and a GPU does.

  • source
  • parent
  • hideshow 4 child comments
  • [–] -1 points 1 year ago (3 children)

    For 2000$ it "claims" to do 140 TOPS of INT8
    When a Intel Core Ultra 7 265K does 33 TOPS of INT8 for 284$

    Don't get me wrong, I would LOVE to buy a chinese GPU at a reasonnable price but this isn't even price competitive with CPUs let alone GPUs.

  • source
  • parent
  • hideshow 3 child comments
  • [–] [S] 9 points 1 year ago (2 children)

    Again, completely different purposes here.

  • source
  • parent
  • hideshow 2 child comments
  • [+] -4 points 1 year ago (16 children)
  • [–] [S] 11 points 1 year ago* (last edited 1 year ago) (15 children)

    that's some quality cope there

  • source
  • parent
  • hideshow 15 child comments
  • [+] -1 points 1 year ago (14 children)
  • [–] [S] 1 point 1 year ago (13 children)

    aww will you look at that, little wasp is mad 🤣

  • source
  • parent
  • hideshow 13 child comments
  • [+] 0 points 1 year ago (12 children)
  • [–] [S] 1 point 1 year ago (11 children)

    I don't need to try harder, you're raging as it is. Don't want you to have a aneurysm.

  • source
  • parent
  • hideshow 11 child comments
  • [+] 1 point 1 year ago (10 children)
  • [–] [S] 2 points 1 year ago (9 children)

    I love how you can't help yourself but keep replying here. Keep on seething there little buddy, it's adorable.

  • source
  • parent
  • hideshow 9 child comments
  • [+] 2 points 1 year ago (8 children)
  • [–] [S] 1 point 1 year ago (7 children)

    I'm just entertained by you losing your shit here. You're just free entertainment for me.

  • source
  • parent
  • hideshow 7 child comments
  • [+] 1 point 1 year ago (6 children)
  • [–] [S] 1 point 1 year ago (5 children)

    so mad, I can just picture you stomping your little feets there 🤣

  • source
  • parent
  • hideshow 5 child comments
  • [+] 1 point 1 year ago (4 children)
  • [–] -2 points 1 year ago

    This product is no longer available.

  • source
  • 1 comment in c/gpu@programming.dev view that post
    [–] 3 points 1 year ago* (last edited 1 year ago)

    Can't wait for the comparison video. If Huawei manages to contend with NVIDIA, it'll flip the market. We need some damn competition.

    Anti Commercial-AI license

  • source