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
That’s not how MoE models work. There are many “expert’ models and there is a static router model (dense) which determines which “expert” models to route the tokens through. What you want at a minimum is the dense portion of the model to be on VRAM and all of the weights to be in RAM/VRAM for best performance.
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)
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.
GGUFs are starting to be available now
Wouldn’t the whole 125B non-n-gram parameters still have to fit into VRAM, though?
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
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
That’s not how MoE models work. There are many “expert’ models and there is a static router model (dense) which determines which “expert” models to route the tokens through. What you want at a minimum is the dense portion of the model to be on VRAM and all of the weights to be in RAM/VRAM for best performance.
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)