github ggml-org/llama.cpp b7820

latest releases: b7822, b7821
4 hours ago
Details

ggml-hexagon: flash-attn opt (#19025)

  • optimize flash attention kernel by improving score computation and online softmax update

  • wip

  • Refactor online softmax update in flash attention kernel for improved performance

  • Optimize flash attention kernel by replacing float array with HVX_Vector for score computation

  • wip

macOS/iOS:

Linux:

Windows:

openEuler:

Don't miss a new llama.cpp release

NewReleases is sending notifications on new releases.