]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
cuda: Q1_0 initial backend (llama/21629)
authorPasha Khosravi <redacted>
Wed, 15 Apr 2026 16:38:38 +0000 (09:38 -0700)
committerGeorgi Gerganov <redacted>
Thu, 30 Apr 2026 08:29:10 +0000 (11:29 +0300)
commitc6d1fbf31f3f8c611772e3a6bb3d3b35ac5f01eb
treea56167ed3312a649abcfcb9e82b32a38811e5872
parent2a785c596944da4cc67d15c8600f606b5021e7bb
cuda: Q1_0 initial backend (llama/21629)

* [cuda] initial Q1_0 backend

* remove unused code, fix AMD MMA guard

* attempt to support dp4a

* Apply suggestions from code review

Co-authored-by: Johannes Gäßler <redacted>
---------

Co-authored-by: Johannes Gäßler <redacted>
ggml/src/ggml-cuda/common.cuh
ggml/src/ggml-cuda/convert.cu
ggml/src/ggml-cuda/dequantize.cuh
ggml/src/ggml-cuda/getrows.cu
ggml/src/ggml-cuda/ggml-cuda.cu
ggml/src/ggml-cuda/mmq.cu
ggml/src/ggml-cuda/mmq.cuh
ggml/src/ggml-cuda/mmvq.cu
ggml/src/ggml-cuda/template-instances/generate_cu_files.py
ggml/src/ggml-cuda/template-instances/mmq-instance-q1_0.cu [new file with mode: 0644]
ggml/src/ggml-cuda/vecdotq.cuh