]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
CUDA: fix FA FP16 accumulator overflow for Granite (#18614)
authorJohannes Gäßler <redacted>
Mon, 5 Jan 2026 18:51:13 +0000 (19:51 +0100)
committerGitHub <redacted>
Mon, 5 Jan 2026 18:51:13 +0000 (19:51 +0100)
commitdf17a4c94f09c0e978e83102fcdbdf6020599460
tree0468118841118a6d4b8fb82d30593b0cd519682a
parent1871f0ba56e57826c1c630c5f57274624d68788e
CUDA: fix FA FP16 accumulator overflow for Granite (#18614)
ggml/src/ggml-cuda/fattn-common.cuh