Tether releases QVAC cross-platform BitNet LoRA framework: supporting training billion-parameter AI models on consumer devices.

According to an official announcement, Tether has released a cross-platform BitNet LoRA tuning framework within QVAC Fabric, enabling optimized training and inference for Microsoft BitNet (1-bit LLM). This framework significantly reduces computational and memory requirements, allowing billion-parameter models to be trained and tuned on laptops, consumer GPUs, and smartphones. This solution is the first to achieve BitNet model tuning on mobile GPUs (including Adreno, Mali, and Apple Bionic). Tests show that a 125M parameter model can be tuned in approximately 10 minutes, a 1B model in about an hour, and even scalable to 13B parameter models on mobile devices. Furthermore, the framework supports heterogeneous hardware such as Intel, AMD, and Apple Silicon, and is the first to achieve 1-bit LLM LoRA tuning on non-NVIDIA devices. In terms of performance, BitNet models achieve 2 to 11 times faster inference speeds on mobile GPUs compared to CPUs, while reducing memory usage by up to 77.8% compared to traditional 16-bit models. Tether stated that this technology has the potential to break the dependence on high-end computing power and cloud infrastructure, promote the development of AI training towards decentralization and localization, and provide a foundation for new application scenarios such as federated learning.