Files
llama_cpp/ggml/src/ggml-cuda/template-instances/fattn-mma-f16-instance-ncols1_4-ncols2_8.cu
T
AesSedaiandGitHub 046e284437 Add flash attention MMA / Tiles to support MiMo-V2.5 (#22812)
* mimo-v2.5: add flash attention mma/tiles for for d_kq=192 d_v=128

* mimo-v2.5: follow (256, 256) fattn templates

* mimo-v2.5: cleanup comments

* mimo-v2.5: further comment cleanup

* mimo-v2.5: address PR feedback
fix GQA handling
check for other dangling 320/576 carveouts and mirror them for 192
Add to backend ops test so new paths are covered
2026-05-09 11:28:29 +08:00

13 lines
439 B
Plaintext

// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-mma-f16.cuh"
DECL_FATTN_MMA_F16_CASE(64, 64, 4, 8);
DECL_FATTN_MMA_F16_CASE(80, 80, 4, 8);
DECL_FATTN_MMA_F16_CASE(96, 96, 4, 8);
DECL_FATTN_MMA_F16_CASE(112, 112, 4, 8);
DECL_FATTN_MMA_F16_CASE(128, 128, 4, 8);
DECL_FATTN_MMA_F16_CASE(192, 128, 4, 8);
DECL_FATTN_MMA_F16_CASE(256, 256, 4, 8);
DECL_FATTN_MMA_F16_CASE(512, 512, 4, 8);