llama.cpp
cb300598 - Feature: Added LIGHTNING_INDEXER support for Deepseek V4 ops on Vulkan Backend (#27453)

Commit
6 days ago
Feature: Added LIGHTNING_INDEXER support for Deepseek V4 ops on Vulkan Backend (#27453) * vulkan: add LIGHTNING_INDEXER op * vulkan: updated lightning_indexer.comp and ggml-vulkan.cpp with 128-lane dot-product reduction moved from a shared-memory tree to subgroupAdd. * vulkan: cleanup; Skip bounds checks * vulkan: cleanup FA_K_ONLY * Revert "vulkan: cleanup FA_K_ONLY" This reverts commit fdcbdd91511945d6878d9070b5447e4a34dce010. * vulkan: restore interleaved K/V buffer ordering * vulkan: Remove FA_K_ONLY * vulkan: Revert flash_attn_dequant * vulkan: Revert tests in backend-ops.cpp
Author
Parents
Loading