Daniel Hiltgen e823bff873 gemma4: enable flash attention (#15378) ...

Backport GGML kernels so we can enable flash attention for the gemma 4 model on
Metal and CUDA.

2026-04-07 08:12:36 -07:00

..

ggml_test.go

ggml: fix crash for array head counts

2025-04-27 11:38:06 -07:00

ggml.go

gemma4: enable flash attention (#15378)

2026-04-07 08:12:36 -07:00

gguf_test.go

ggml: ensure tensor size is valid (#14406)

2026-02-24 21:52:44 -04:00

gguf.go

ggml: ensure tensor size is valid (#14406)

2026-02-24 21:52:44 -04:00

type.go

fs/ggml: fix function name in comment (#12630)

2025-10-15 21:53:38 -07:00