ggml/llama.cpp releases: b11986

TL;DR

The ggml/llama.cpp project added initial OpenCL support for IQ2, enhancing matrix multiplication methods specifically for Adreno GPUs. This update is crucial as it broadens the library's compatibility and performance on mobile devices with Adreno hardware.

Detailed Summary

The ggml/llama.cpp project has released updates that include initial support for IQ2 on OpenCL, enhancing matrix multiplication methods specifically for Adreno GPUs. Co-authored by Hongqiang, these changes add and optimize various matrix operations such as AOS and SOA mv, tiled mm, dp4a, and narrow batches decode. This update broadens the library's compatibility with mobile and embedded devices using Adreno GPUs, improving performance in computational tasks.

Key Points

• • opencl: initial support for iq2 (#31267)
• • opencl: add iq2_s AOS mv
• • opencl: add iq2_s SOA mv and tiled mm