Topic: tq1_0

The ggml/llama.cpp project added support for TQ1_0 in Vulkan, enhancing matrix operations and dequantization processes. This update is significant as it optimizes memory usage and performance for quantized models.