rohanpaul_ai· @rohanpaul_ai · X·· 1 天前AI 评分
Qwen 27B 1.5bit量化仅需6GB
Emad Mostaque：AI 模型"基本已达到人脑的效率"。 Alibaba 的 Qwen 27B 通常需要约 60GB 内存，而 Prism ML 团队通过用 1.5 bit 而非 16 bit 存储数值，将其降至 6GB，同时保留 95% 的性能。 在 6GB 下它可在 Raspberry Pi 上运行，"其能耗与你的大脑相当。" ---- 完整视频在 "Tom Bilyeu" YouTube 频道，（链接在评论中）
Emad Mostaque: AI models "have pretty much reached the efficiency of a human brain".
Alibaba's Qwen 27B normally needs about 60GB of memory, and a team from Prism ML cut that to 6GB by storing its numbers with 1.5 bits instead of 16 while keeping 95% of its performance.
At 6GB it runs on a Raspberry Pi, "which uses as much energy as your brain."
----
Full video on "Tom Bilyeu" YouTube channel, (link in comment)
来源：rohanpaul_ai · x.com