Footnote on page 3.
>Total parameter count governs storage: at FP4, MIXTRAL-8X7B fits within 24 GB GDDR6 (Quadro RTX 6000), and GPT-OSS-120B fits within 128 GB unified memory (Apple M4 Max).
>Total parameter count governs storage: at FP4, MIXTRAL-8X7B fits within 24 GB GDDR6 (Quadro RTX 6000), and GPT-OSS-120B fits within 128 GB unified memory (Apple M4 Max).
No comments yet.