Commit d9cc25b
- .github/actions/windows-setup-cuda/action.yml+31
- .github/workflows/build-3rd-party.yml+3-3
- .github/workflows/build-and-test-snapdragon.yml+2-2
- .github/workflows/build-android.yml+53-3
- .github/workflows/build-apple.yml+82-10
- .github/workflows/build-cache.yml+4-4
- .github/workflows/build-cann.yml+71-69
- .github/workflows/build-cross.yml+1-1
- .github/workflows/build-hip.yml+167
- .github/workflows/build-ibm.yml+150
- .github/workflows/build-msys.yml+3-3
- .github/workflows/build-opencl.yml+83
- .github/workflows/build-openvino.yml+4-4
- .github/workflows/build-riscv.yml+83-9
- .github/workflows/build-rpc.yml+67
- .github/workflows/build-sanitize.yml+37-25
- .github/workflows/build-self-hosted.yml+56-7
- .github/workflows/build-sycl.yml+128-126
- .github/workflows/build-vulkan.yml+4-4
- .github/workflows/build-webgpu.yml+173
- .github/workflows/build.yml+63-657
- .github/workflows/hip-quality-check.yml+3-3
- .github/workflows/release.yml+360-314
- .github/workflows/server-sanitize.yml+18-18
- .github/workflows/server-self-hosted.yml+4-4
- .github/workflows/server.yml+26-23
- .github/workflows/ui-self-hosted.yml+4-4
- .github/workflows/ui.yml+4-4
- CONTRIBUTING.md+1
- ci/run.sh+7-6
- common/arg.cpp+5-5
- conversion/__init__.py+2
- conversion/base.py+6
- conversion/gemma.py+1-1
- conversion/talkie.py+53
- convert_hf_to_gguf_update.py+2
- convert_lora_to_gguf.py+10
- docs/autoparser.md+1-1
- docs/backend/SYCL.md+2
- docs/backend/snapdragon/README.md+1-1
- docs/multimodal/granitevision.md+1-1
- examples/model-conversion/README.md+1-1
- ggml/src/ggml-cuda/common.cuh+6-3
- ggml/src/ggml-cuda/fwht.cu+14-21
- ggml/src/ggml-cuda/fwht.cuh+2-1
- ggml/src/ggml-cuda/ggml-cuda.cu+1-3
- ggml/src/ggml-hexagon/ggml-hexagon.cpp+24
- ggml/src/ggml-hexagon/htp/CMakeLists.txt+1
- ggml/src/ggml-hexagon/htp/concat-ops.c+275
- ggml/src/ggml-hexagon/htp/cpy-ops.c+165-145
- ggml/src/ggml-hexagon/htp/get-rows-ops.c+107-13
- ggml/src/ggml-hexagon/htp/hmx-flash-attn-ops.c+11-10
- ggml/src/ggml-hexagon/htp/htp-ctx.h+1
- ggml/src/ggml-hexagon/htp/htp-ops.h+1
- ggml/src/ggml-hexagon/htp/hvx-sin-cos.h+90
- ggml/src/ggml-hexagon/htp/hvx-utils.h+2
- ggml/src/ggml-hexagon/htp/main.c+4-2
- ggml/src/ggml-hexagon/htp/rope-ops.c+180-60
- ggml/src/ggml-hexagon/htp/set-rows-ops.c+6
- ggml/src/ggml-hexagon/htp/unary-ops.c+1-1
- ggml/src/ggml-metal/ggml-metal-device.h+26
- ggml/src/ggml-metal/ggml-metal-device.m+46
- ggml/src/ggml-sycl/common.hpp+3
- ggml/src/ggml-sycl/ggml-sycl.cpp+160-11
- ggml/src/ggml-vulkan/ggml-vulkan.cpp+114-13
- ggml/src/ggml-vulkan/vulkan-shaders/conv2d_mm.comp+146-13
- ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp+10-2
- ggml/src/ggml-webgpu/ggml-webgpu-shader-lib.hpp+112-119
- ggml/src/ggml-webgpu/ggml-webgpu.cpp+127-73
- ggml/src/ggml-webgpu/wgsl-shaders/common_decls.tmpl+1-2
- ggml/src/ggml-webgpu/wgsl-shaders/mul_mat.wgsl-747
- ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_vec.wgsl+20-2
- ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_vec_acc.tmpl-1
- ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_vec_q_acc.tmpl+303
- ggml/src/ggml-webgpu/wgsl-shaders/quantize_q8.wgsl+173
- ggml/src/ggml-zendnn/ggml-zendnn.cpp+5-5
- gguf-py/gguf/constants.py+15
- gguf-py/gguf/tensor_mapping.py+10
- gguf-py/pyproject.toml+1-1
- pyproject.toml+12-9
- scripts/sync_vendor.py+1-1
- src/llama-arch.cpp+1
- src/llama-arch.h+1
- src/llama-model.cpp+3
- src/llama-model.h+1-1
- src/llama-vocab.cpp+13-1
- src/llama-vocab.h+1
- src/models/mistral3.cpp+8-4
- src/models/models.h+13
- src/models/talkie.cpp+149
- tests/gguf-model-data.cpp+10-8
- tests/test-backend-ops.cpp+138-45
- tools/cli/README.md+6-5
- tools/completion/README.md+4-4
- tools/server/README.md+6-5
- tools/server/server-http.cpp+3-2
- tools/server/server-http.h+1
- vendor/cpp-httplib/httplib.cpp+250-42
- vendor/cpp-httplib/httplib.h+40-5
0 commit comments