ggml/llama.cpp releases: b11868

TL;DR

The ggml/llama.cpp project updated its codebase to include a split KV-head team for multicore HMX flash attention, enhancing parallel processing capabilities. This update is crucial as it improves the efficiency and performance of large language models across multiple cores.

Detailed Summary

The ggml/llama.cpp project updated its codebase to include a split of the KV-head team for multicore HMX flash attention, with changes merged into a single GGML_HEXAGON_FA_SPLIT flag. This update affects how attention is handled in parallel processing environments, potentially improving performance and efficiency. The broader impact could be enhanced capabilities for large language models running on multi-core hardware.

Key Points

• • hexagon: KV-head team split for HMX flash attention in multicore
• • hexagon: merge FA_HEAD_SPLIT and FA_SPLIT into single GGML_HEXAGON_FA_SPLIT flag