Why Is NVIDIA's 4-Bit Qwen 3.8 27B a 21.9 GB Download, Not 13.9?
中文摘要
本文解释了为何 NVIDIA 的 4-bit Qwen 3.8 27B 模型大小为 21.9 GB 而非预期的 13.9 GB。
English Summary
This article explains why NVIDIA's 4-bit Qwen 3.8 27B model is 21.9 GB instead of the expected 13.9 GB.
原文节选
Continue reading on Towards AI »