]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
CUDA: fix crash on large batch size for quant. MoE (llama/13537)
authorJohannes Gäßler <redacted>
Wed, 14 May 2025 14:41:02 +0000 (16:41 +0200)
committerGeorgi Gerganov <redacted>
Mon, 19 May 2025 11:58:39 +0000 (14:58 +0300)
commit0dda27bc0bbd761fb2dd216d1a32bab660a42cda
tree11ce9a8aa4c4dcc07fb8037e4db87c8751be6a16
parentffa4720f25837355b4b0f3498596ebf6cd776395
CUDA: fix crash on large batch size for quant. MoE (llama/13537)
ggml/src/ggml-cuda/mmq.cu
ggml/src/ggml-cuda/quantize.cu