this post was submitted on 26 Aug 2026
43 points (100.0% liked)

LocalLLaMA

5125 readers
33 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
MODERATORS
top 11 comments
sorted by: hot top controversial new old
[–] Asafum@lemmy.world 5 points 1 week ago

I heard an MoE was coming but its a bummer to see it so large. 64gb ram and 32vram, but I'm stuck with a shitbag Intel card that needs mirrored vram to ram so I can't have enough of this offloaded. Hopefully the supposed stealth model announced today is the qwen4 MoE in the 35b range :/

[–] BeefAndPoultry@lemmus.org 5 points 1 week ago* (last edited 1 week ago) (3 children)

Number of Parameters: 125B with 6B activated, plus 51B n-gram embedding and 4B MTP

I bet you could run this in Q4 on 96GB RAM and 16GB VRAM, maybe even less. The benchmark scores seem good, beating 27b and DeepSeek Flash.

The n-gram embeddings sound very similar to Gemma 4 e4b embeddings. Need llama.cpp to support streaming n-grams from SSD, mmap would be less efficient than having explicit support.

GGUFs are starting to be available now

[–] e0qdk@reddthat.com 2 points 1 week ago

I'm downloading the full weights currently. Will give it a try on my Framework Desktop when I can. I expect that I will need a newer version of llama.cpp given the new architecture -- there are already pull requests pending though... (e.g. https://github.com/ggml-org/llama.cpp/pull/27742)

[–] ZephyrXero@lemmy.world 2 points 1 week ago

I've gotten 35B-A3B running on a 4GB GPU with 16GB of RAM, the 6B actives should easily fit onto an 8GB card, but yeah, you might could get away with just 64GB of memory, maybe even just 32 if quantized small enough

[–] Multiplexer@discuss.tchncs.de 1 points 1 week ago (1 children)

Wouldn't the whole 125B non-n-gram parameters still have to fit into VRAM, though?

[–] BeefAndPoultry@lemmus.org 5 points 1 week ago* (last edited 1 week ago)

It's MoE so you can use --n-cpu-moe

https://lemmus.org/post/24235317

Low number of active parameters (6B) means you don't need much VRAM to get decent speeds

[–] panda_abyss@lemmy.ca 4 points 1 week ago (1 children)

I’m pretty excited about this one, is it supported by llama.cpp main yet?

[–] e0qdk@reddthat.com 4 points 1 week ago (1 children)

Not on main yet I think (as of ~2:50PM UTC on 2026-08-26) -- there's a link to the PR for it in my other comment though. Unsloth's fork has that integrated (they submitted the PR). I wouldn't be surprised if something lands quickly in main, but this is a new architecture so may take a bit for people to figure out how to get the most out of it -- bunch of discussion about e.g. SSD offloading for the ngrams and stuff like that in the github thread.

[–] SirDimples@programming.dev 2 points 1 week ago* (last edited 1 week ago) (1 children)

llama.cpp support for it got merged a few hours ago 🏆

[–] e0qdk@reddthat.com 2 points 1 week ago

Nice! Hopefully I can figure out how to actually get it to load tomorrow... (It keeps getting OOM-killed when I try.)

[–] SirDimples@programming.dev 2 points 1 week ago* (last edited 1 week ago)

I've been testing this model since yesterday through the Qwen API in opencode and deepseek harness. I actually prefer it over GLM-5.3-Flash! it's super good at agentic coding and comfortably fast, now my goal is to be able to one day run it locally 😳 I think a Strix Halo with 128GB and a fast nvme should be able to run it well at near loseless quant