[sourcecodeplz]

With RAM you would need at least 500gb to load it but some 100-200gb more for context too. Pair it with a 24gb GPU and the speed will be 10t/s, at least, I estimate.

[danielhanchen]

Oh yes for the FP8, you will need 500GB ish. 4bit around 250GB - offloading MoE experts / layers to RAM will definitely help - as you mentioned a 24GB card should be enough!

[vFunct]

Do we know if the full model is FP8 or FP16/BF16? The hugging face page says BF16: https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct

So likely it needs 2x the memory.