PrismML 把 Qwen3.6-27B 压到 5.9GB,手机都能本地跑,还全开源,适合离线使用。
PrismML 推出了 Bonsai 27B,这是对 Qwen3.6-27B 进行低比特量化得到的模型,不修改原架构。三元版本使用 {−1, 0, +1} 权重,每个参数仅 1.71 bits,理想大小 5.9GB。1-bit 版本采用二进制 {−1, +1} 权重。两个版本均以 Apache 2.0 许可开源,可在笔记本和手机上本地运行。
PrismML Releases Bonsai 27B: 1-bit and Ternary Builds of Qwen3.6-27B That Run on Laptops and Phones
PrismML just released Bonsai 27B. It is a low-bit representation of Qwen3.6-27B, not a new pretrain. The architecture is unchanged. Two variants ship under Apache 2.0. Ternary Bonsai 27B uses {−1, 0, +1} weights at a true 1.71 bits per weight. Its ideal size is 5.9GB. 1-bit Bonsai 27B uses binary {−1, +1} weights at […] The post PrismML Releases Bonsai 27B: 1-bit and Ternary Builds of Qwen3.6-27B That Run on Laptops and Phones appeared first on MarkTechPost .