this post was submitted on 10 Oct 2026

26 points (86.1% liked)

LocalLLaMA

localllama@sh.itjust.works

5,609 readers

3,211 active

873 users here now

Welcome to LocalLLaMA! Here we discuss running and developing machine learning models at home. Lets explore cutting edge open source neural network technology together.

Get support from the community! Ask questions, share prompts, discuss benchmarks, get hyped at the latest and greatest model releases! Enjoy talking about our awesome hobby.

As ambassadors of the self-hosting machine learning community, we strive to support each other and share our enthusiasm in a positive constructive way.

Rules:

Rule 1 - No harassment or personal character attacks of community members. I.E no namecalling, no generalizing entire groups of people that make up our community, no baseless personal insults.

Rule 2 - No comparing artificial intelligence/machine learning models to cryptocurrency. I.E no comparing the usefulness of models to that of NFTs, no comparing the resource usage required to train a model is anything close to maintaining a blockchain/ mining for crypto, no implying its just a fad/bubble that will leave people with nothing of value when it burst.

Rule 3 - No comparing artificial intelligence/machine learning to simple text prediction algorithms. I.E statements such as "llms are basically just simple text predictions like what your phone keyboard autocorrect uses, and they're still using the same algorithms since <over 10 years ago>.

Rule 4 - No implying that models are devoid of purpose or potential for enriching peoples lives.

founded 3 years ago

moderation log

moderators

MonsterBug@sh.itjust.works

Smokeydope@lemmy.world

pax@sh.itjust.works

noneabove1182@sh.itjust.works

What Qwen4 sizes are you hoping for? (lemmy.ml)

submitted 1 month ago by robber@lemmy.ml to c/localllama@sh.itjust.works

13 comments fedilink hide all child comments

Basically the title.

For me, a dense model around 20b, with an additional engram table of 10-20b, or a sparse MoE around 80b would be amazing.

all 13 comments

sorted by: hot top controversial new old

[–] dosse91@lemmy.trippy.pizza 11 points 1 month ago (1 child)

Flash next 125b is genuinely amazing, I hope to see 4.0 of the same size.

permalink

fedilink

source

hideshow 1 child comment

[–] tpWinthropeIII@lemmy.world 3 points 1 month ago

Agree, and upvoted.

Another good size would be 160 gb for 96 gb vram systems. After that I would like to see 225 to 250 gb.

permalink

fedilink

source

parent

[–] BeefAndPoultry@lemmus.org 7 points 1 month ago

32GB RAM and 8GB VRAM here, so 35b a3b is about my maximum, maybe with added n-grams

permalink

fedilink

source

[–] fhein@lemmy.world 5 points 1 month ago (3 children)

27B dense was a good size for 3.8, but I'm guessing they're going to add a few B parameters in order to make it noticeably better.

permalink

fedilink

source

hideshow 3 child comments

[–] BeefAndPoultry@lemmus.org 2 points 1 month ago*

If they can make the KV cache size scale more efficiently with many tokens, like DeepSeek 4.1 Flash fitting 1 million tokens in 890MB, it might give them enough room to add a couple billion parameters of weights and still fit well in 32GB GPUs

permalink

fedilink

source

parent

[–] robber@lemmy.ml [S] 2 points 1 month ago (1 child)

Not necessarly, I'd say. We've seen many releases with multiple sizes and the 27B class has been around since at least the Gemma3 times.

permalink

fedilink

source

parent

hideshow 1 child comment

[–] fhein@lemmy.world 4 points 1 month ago

Well, it's just a guess. But also Gemma4 increased to 31B, so I think there's a possibility that Qwen does the same.

permalink

fedilink

source

parent

[–] e0qdk@reddthat.com 13 points 1 month ago (1 child)

Another 35B-A3B MoE or similar would be nice...

FYI, Yandex released an 80B MoE base model (i.e. not instruction tuned!) recently: https://huggingface.co/yandex/AliceAI-Foundation-80B-A3B-Base -- haven't experimented with it personally though.

permalink

fedilink

source

hideshow 1 child comment