this post was submitted on 02 Feb 2025
57 points (79.4% liked)

Technology

63186 readers
3492 users here now

This is a most excellent place for technology news and articles.


Our Rules


  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each other!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed
  10. Accounts 7 days and younger will have their posts automatically removed.

Approved Bots


founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] Paradox@lemdro.id 129 points 3 weeks ago (11 children)

Can I download their model and run it on my own hardware? No? Then they're inferior to deepseek

[–] Teanut@lemmy.world 49 points 3 weeks ago (10 children)

In fairness, unless you have about 800GB of VRAM/HBM you're not running true Deepseek yet. The smaller models are Llama or Qwen distilled from Deepseek R1.

I'm really hoping Deepseek releases smaller models that I can fit on a 16GB GPU and try at home.

[–] brucethemoose@lemmy.world 14 points 3 weeks ago* (last edited 3 weeks ago) (6 children)

Qwen 2.5 is already amazing for a 14B, so I don’t see how deepseek can improve that much with a new base model, even if they continue train it.

Perhaps we need to meet in the middle, and have quad channel APUs like Strix Halo become more common, and maybe release like 40-80GB MoE models. Perhaps bitnet ones?

Or design them for asynchronous inference.

I just don’t see how 20B-ish models can perform like one orders of magnitude bigger without a paradigm shift.

[–] Teanut@lemmy.world 7 points 3 weeks ago (2 children)

nVidia's new Digits workstation, while expensive from a consumer standpoint, should be a great tool for local inferencing research. $3000 for 128GB isn't a crazy amount for a university or other researcher to spend, especially when you look at the price of the 5090.

[–] Viri4thus@feddit.org 3 points 3 weeks ago

Why would you buy a single use behemoth when you can buy a strix halo 128GB that can work as an actual tablet/laptop and have all the functionality of the behemoth?! while supporting decades of legacy x86 software. Truly wondering why anyone would buy that NVIDIA thing other than pure ignorance and marketing says NV is the AI company.

[–] brucethemoose@lemmy.world 1 points 3 weeks ago

Dense models that would fit in 100-ish GB like mistral large would be really slow on that box, and there isn't a SOTA MoE for that size yet.

So, unless you need tons of batching/parallel requests, its... kinda neither here nor there?

As someone else said, the calculus changes with cheaper Strix Halo boxes (assuming those mini PCs are under $3K).

load more comments (3 replies)
load more comments (6 replies)
load more comments (6 replies)