PrismML hopes its tiny LLM will change how we all use AI
TechCrunch
Read full postPrismML released Bonsai 2 27B, a compressed version of Alibaba's Qwen3.8 27B model, reducing size from about 59GB to 5.9GB with minimal performance loss, enabling use on PCs and smartphones. The startup aims to make high-performing LLMs accessible on smaller devices without significant accuracy trade-offs.




