| ggml/include/ggml-opt.h | 18 ++ | |
| ggml/src/ggml-cpu/ops.cpp | 80 +++--- | |
| ggml/src/ggml-cuda/ggml-cuda.cu | 38 ++- | |
| ggml/src/ggml-cuda/out-prod.cu | 602 +++++++++++++++++++++++++++++++++++---- | |
| ggml/src/ggml-cuda/unary.cu | 81 +++++- | |
| ggml/src/ggml-opt.cpp | 330 ++++++++++++++++++++- | |
| ggml/src/ggml.c | 403 +++++++++++++++++++++++++- | |
| include/llama.h | 27 ++ | |
| src/llama-adapter.cpp | 24 ++ | |
| src/llama-context.cpp | 616 ++++++++++++++++++++++++++++++++++++++++ | |
| src/llama-context.h | 15 + | |
| src/llama-graph.cpp | 190 ++++++++++--- | |
| src/models/delta-net-base.cpp | 8 +- | |
| src/models/qwen35.cpp | 73 ++++- | |
| tests/CMakeLists.txt | 39 +++ | |
| tests/test-backend-ops.cpp | 124 ++++++++ | |
| 16 files changed, 2507 insertions(+), 161 deletions(-) | |