PrismML releases Bonsai 2 27B, which compresses Alibaba's Qwen3.8 27B to 5.9 GB, small enough for smartphones, while retaining 98.2% of Qwen's benchmark scores (Julie Bort/TechCrunch)

PrismML has released Bonsai 2 27B, a new model that compresses Alibaba's Qwen3.8 27B to a mere 5.9 GB. This significant reduction in size makes it suitable for smartphones. Despite the compression, Bonsai 2 27B remarkably retains 98.2% of Qwen's original benchmark scores, indicating its efficiency and performance. This development by PrismML could be a notable advancement in making powerful AI models more accessible on mobile devices.