@novastera-oss/llamarn 0.3.0 → 0.3.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/android/build.gradle +2 -1
- package/android/proguard-rules.pro +12 -0
- package/android/src/main/cpp/include/llama.h +15 -47
- package/android/src/main/jniLibs/arm64-v8a/libggml-base.so +0 -0
- package/android/src/main/jniLibs/arm64-v8a/libggml-cpu.so +0 -0
- package/android/src/main/jniLibs/arm64-v8a/libggml.so +0 -0
- package/android/src/main/jniLibs/arm64-v8a/libllama.so +0 -0
- package/android/src/main/jniLibs/armeabi-v7a/libggml-base.so +0 -0
- package/android/src/main/jniLibs/armeabi-v7a/libggml-cpu.so +0 -0
- package/android/src/main/jniLibs/armeabi-v7a/libggml.so +0 -0
- package/android/src/main/jniLibs/armeabi-v7a/libllama.so +0 -0
- package/android/src/main/jniLibs/x86/libggml-base.so +0 -0
- package/android/src/main/jniLibs/x86/libggml-cpu.so +0 -0
- package/android/src/main/jniLibs/x86/libggml.so +0 -0
- package/android/src/main/jniLibs/x86/libllama.so +0 -0
- package/android/src/main/jniLibs/x86_64/libggml-base.so +0 -0
- package/android/src/main/jniLibs/x86_64/libggml-cpu.so +0 -0
- package/android/src/main/jniLibs/x86_64/libggml.so +0 -0
- package/android/src/main/jniLibs/x86_64/libllama.so +0 -0
- package/cpp/build-info.cpp +2 -2
- package/cpp/llama.cpp/CMakePresets.json +11 -0
- package/cpp/llama.cpp/CODEOWNERS +1 -0
- package/cpp/llama.cpp/README.md +4 -3
- package/cpp/llama.cpp/common/arg.cpp +45 -1
- package/cpp/llama.cpp/common/common.cpp +22 -6
- package/cpp/llama.cpp/common/common.h +18 -4
- package/cpp/llama.cpp/convert_hf_to_gguf.py +500 -32
- package/cpp/llama.cpp/convert_hf_to_gguf_update.py +12 -13
- package/cpp/llama.cpp/ggml/CMakeLists.txt +6 -1
- package/cpp/llama.cpp/ggml/cmake/ggml-config.cmake.in +85 -47
- package/cpp/llama.cpp/ggml/include/ggml-webgpu.h +19 -0
- package/cpp/llama.cpp/ggml/src/CMakeLists.txt +1 -0
- package/cpp/llama.cpp/ggml/src/ggml-alloc.c +0 -15
- package/cpp/llama.cpp/ggml/src/ggml-backend-reg.cpp +7 -0
- package/cpp/llama.cpp/ggml/src/ggml-backend.cpp +8 -20
- package/cpp/llama.cpp/ggml/src/ggml-cann/acl_tensor.cpp +3 -1
- package/cpp/llama.cpp/ggml/src/ggml-cann/aclnn_ops.cpp +58 -3
- package/cpp/llama.cpp/ggml/src/ggml-cann/aclnn_ops.h +130 -22
- package/cpp/llama.cpp/ggml/src/ggml-cann/ggml-cann.cpp +122 -16
- package/cpp/llama.cpp/ggml/src/ggml-cpu/CMakeLists.txt +5 -2
- package/cpp/llama.cpp/ggml/src/ggml-cpu/arch/loongarch/quants.c +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-cpu/kleidiai/kernels.cpp +109 -12
- package/cpp/llama.cpp/ggml/src/ggml-cpu/kleidiai/kernels.h +3 -0
- package/cpp/llama.cpp/ggml/src/ggml-cpu/kleidiai/kleidiai.cpp +88 -10
- package/cpp/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp +343 -1094
- package/cpp/llama.cpp/ggml/src/ggml-cpu/ops.cpp +3 -0
- package/cpp/llama.cpp/ggml/src/ggml-cpu/repack.cpp +0 -1
- package/cpp/llama.cpp/ggml/src/ggml-cpu/vec.cpp +3 -0
- package/cpp/llama.cpp/ggml/src/ggml-cuda/CMakeLists.txt +3 -3
- package/cpp/llama.cpp/ggml/src/ggml-cuda/common.cuh +14 -4
- package/cpp/llama.cpp/ggml/src/ggml-cuda/convert.cu +64 -17
- package/cpp/llama.cpp/ggml/src/ggml-cuda/cpy-utils.cuh +225 -0
- package/cpp/llama.cpp/ggml/src/ggml-cuda/cpy.cu +41 -301
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn-common.cuh +85 -67
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn-mma-f16.cuh +45 -62
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn-tile-f16.cu +28 -43
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn-tile-f32.cu +41 -56
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn-vec-f16.cuh +36 -47
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn-vec-f32.cuh +31 -43
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn-wmma-f16.cu +22 -37
- package/cpp/llama.cpp/ggml/src/ggml-cuda/fattn.cu +3 -13
- package/cpp/llama.cpp/ggml/src/ggml-cuda/ggml-cuda.cu +73 -23
- package/cpp/llama.cpp/ggml/src/ggml-cuda/im2col.cu +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-cuda/mma.cuh +111 -3
- package/cpp/llama.cpp/ggml/src/ggml-cuda/mmq.cu +6 -4
- package/cpp/llama.cpp/ggml/src/ggml-cuda/mmq.cuh +1152 -689
- package/cpp/llama.cpp/ggml/src/ggml-cuda/norm.cu +92 -5
- package/cpp/llama.cpp/ggml/src/ggml-cuda/norm.cuh +2 -0
- package/cpp/llama.cpp/ggml/src/ggml-cuda/set-rows.cu +275 -0
- package/cpp/llama.cpp/ggml/src/ggml-cuda/set-rows.cuh +7 -0
- package/cpp/llama.cpp/ggml/src/ggml-cuda/unary.cu +7 -0
- package/cpp/llama.cpp/ggml/src/ggml-cuda/unary.cuh +2 -0
- package/cpp/llama.cpp/ggml/src/ggml-cuda/vendors/hip.h +13 -1
- package/cpp/llama.cpp/ggml/src/ggml-cuda/vendors/musa.h +2 -2
- package/cpp/llama.cpp/ggml/src/ggml-impl.h +16 -0
- package/cpp/llama.cpp/ggml/src/ggml-metal/ggml-metal-impl.h +13 -3
- package/cpp/llama.cpp/ggml/src/ggml-metal/ggml-metal.m +407 -69
- package/cpp/llama.cpp/ggml/src/ggml-metal/ggml-metal.metal +380 -83
- package/cpp/llama.cpp/ggml/src/ggml-musa/CMakeLists.txt +18 -4
- package/cpp/llama.cpp/ggml/src/ggml-opencl/CMakeLists.txt +2 -0
- package/cpp/llama.cpp/ggml/src/ggml-opencl/ggml-opencl.cpp +295 -2
- package/cpp/llama.cpp/ggml/src/ggml-opencl/kernels/conv2d.cl +185 -0
- package/cpp/llama.cpp/ggml/src/ggml-opencl/kernels/conv2d_f16_f32.cl +176 -0
- package/cpp/llama.cpp/ggml/src/ggml-opencl/kernels/im2col_f16.cl +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-opencl/kernels/im2col_f32.cl +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-opencl/kernels/rms_norm.cl +79 -0
- package/cpp/llama.cpp/ggml/src/ggml-rpc/ggml-rpc.cpp +4 -4
- package/cpp/llama.cpp/ggml/src/ggml-sycl/gemm.hpp +14 -26
- package/cpp/llama.cpp/ggml/src/ggml-sycl/ggml-sycl.cpp +131 -46
- package/cpp/llama.cpp/ggml/src/ggml-sycl/im2col.cpp +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-sycl/quants.hpp +8 -9
- package/cpp/llama.cpp/ggml/src/ggml-sycl/set_rows.cpp +43 -43
- package/cpp/llama.cpp/ggml/src/ggml-sycl/vecdotq.hpp +2 -6
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/ggml-vulkan.cpp +287 -22
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/conv2d_mm.comp +265 -0
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/copy_to_quant.comp +1 -5
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/dequant_q2_k.comp +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/dequant_q3_k.comp +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/dequant_q4_k.comp +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/dequant_q5_k.comp +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/dequant_q6_k.comp +1 -1
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/generic_binary_head.comp +2 -0
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/glu_head.comp +2 -0
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/im2col.comp +3 -8
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rms_norm.comp +8 -2
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_head.comp +1 -4
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rte.comp +5 -0
- package/cpp/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp +71 -16
- package/cpp/llama.cpp/ggml/src/ggml-webgpu/CMakeLists.txt +54 -0
- package/cpp/llama.cpp/ggml/src/ggml-webgpu/ggml-webgpu.cpp +907 -0
- package/cpp/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/cpy.wgsl +60 -0
- package/cpp/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/embed_wgsl.py +35 -0
- package/cpp/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/memset.wgsl +40 -0
- package/cpp/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/mul_mat.wgsl +56 -0
- package/cpp/llama.cpp/ggml/src/ggml.c +4 -6
- package/cpp/llama.cpp/gguf-py/gguf/constants.py +98 -0
- package/cpp/llama.cpp/gguf-py/gguf/metadata.py +4 -0
- package/cpp/llama.cpp/gguf-py/gguf/scripts/gguf_dump.py +24 -1
- package/cpp/llama.cpp/gguf-py/gguf/tensor_mapping.py +75 -52
- package/cpp/llama.cpp/include/llama.h +15 -7
- package/cpp/llama.cpp/models/templates/llama-cpp-rwkv-world.jinja +34 -0
- package/cpp/llama.cpp/models/templates/moonshotai-Kimi-K2.jinja +43 -0
- package/cpp/llama.cpp/requirements/requirements-all.txt +1 -0
- package/cpp/llama.cpp/requirements/requirements-server-bench.txt +5 -0
- package/cpp/llama.cpp/src/llama-arch.cpp +106 -0
- package/cpp/llama.cpp/src/llama-arch.h +5 -0
- package/cpp/llama.cpp/src/llama-batch.cpp +76 -70
- package/cpp/llama.cpp/src/llama-batch.h +24 -18
- package/cpp/llama.cpp/src/llama-chat.cpp +43 -1
- package/cpp/llama.cpp/src/llama-chat.h +2 -0
- package/cpp/llama.cpp/src/llama-context.cpp +180 -106
- package/cpp/llama.cpp/src/llama-context.h +26 -16
- package/cpp/llama.cpp/src/llama-cparams.h +3 -2
- package/cpp/llama.cpp/src/llama-graph.cpp +203 -39
- package/cpp/llama.cpp/src/llama-graph.h +147 -72
- package/cpp/llama.cpp/src/llama-hparams.cpp +40 -0
- package/cpp/llama.cpp/src/llama-hparams.h +10 -2
- package/cpp/llama.cpp/src/llama-kv-cache-unified-iswa.cpp +11 -5
- package/cpp/llama.cpp/src/llama-kv-cache-unified-iswa.h +3 -0
- package/cpp/llama.cpp/src/llama-kv-cache-unified.cpp +698 -302
- package/cpp/llama.cpp/src/llama-kv-cache-unified.h +89 -31
- package/cpp/llama.cpp/src/llama-memory-hybrid.cpp +1 -0
- package/cpp/llama.cpp/src/llama-memory-recurrent.cpp +16 -1
- package/cpp/llama.cpp/src/llama-model.cpp +1293 -312
- package/cpp/llama.cpp/src/llama-model.h +3 -4
- package/cpp/llama.cpp/src/llama-quant.cpp +1 -2
- package/cpp/llama.cpp/src/llama-vocab.cpp +363 -8
- package/cpp/llama.cpp/src/llama-vocab.h +2 -0
- package/cpp/llama.cpp/src/unicode.cpp +207 -0
- package/cpp/llama.cpp/src/unicode.h +2 -0
- package/ios/include/common.h +18 -4
- package/ios/include/llama.h +15 -7
- package/ios/libs/llama.xcframework/Info.plist +15 -15
- package/ios/libs/llama.xcframework/ios-arm64/dSYMs/llama.dSYM/Contents/Resources/DWARF/llama +0 -0
- package/ios/libs/llama.xcframework/ios-arm64/dSYMs/llama.dSYM/Contents/Resources/Relocations/aarch64/llama.yml +5267 -5059
- package/ios/libs/llama.xcframework/ios-arm64/llama.framework/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/ios-arm64/llama.framework/llama +0 -0
- package/ios/libs/llama.xcframework/ios-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/DWARF/llama +0 -0
- package/ios/libs/llama.xcframework/ios-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/Relocations/aarch64/llama.yml +5238 -5030
- package/ios/libs/llama.xcframework/ios-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/Relocations/x86_64/llama.yml +4014 -3889
- package/ios/libs/llama.xcframework/ios-arm64_x86_64-simulator/llama.framework/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/ios-arm64_x86_64-simulator/llama.framework/llama +0 -0
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/dSYMs/llama.dSYM/Contents/Resources/DWARF/llama +0 -0
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/dSYMs/llama.dSYM/Contents/Resources/Relocations/aarch64/llama.yml +5238 -5030
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/dSYMs/llama.dSYM/Contents/Resources/Relocations/x86_64/llama.yml +4016 -3891
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/llama.framework/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/llama.framework/Versions/A/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/llama.framework/Versions/A/llama +0 -0
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/llama.framework/Versions/Current/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/llama.framework/Versions/Current/llama +0 -0
- package/ios/libs/llama.xcframework/macos-arm64_x86_64/llama.framework/llama +0 -0
- package/ios/libs/llama.xcframework/tvos-arm64/dSYMs/llama.dSYM/Contents/Resources/DWARF/llama +0 -0
- package/ios/libs/llama.xcframework/tvos-arm64/dSYMs/llama.dSYM/Contents/Resources/Relocations/aarch64/llama.yml +5267 -5059
- package/ios/libs/llama.xcframework/tvos-arm64/llama.framework/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/tvos-arm64/llama.framework/llama +0 -0
- package/ios/libs/llama.xcframework/tvos-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/DWARF/llama +0 -0
- package/ios/libs/llama.xcframework/tvos-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/Relocations/aarch64/llama.yml +5238 -5030
- package/ios/libs/llama.xcframework/tvos-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/Relocations/x86_64/llama.yml +4014 -3889
- package/ios/libs/llama.xcframework/tvos-arm64_x86_64-simulator/llama.framework/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/tvos-arm64_x86_64-simulator/llama.framework/llama +0 -0
- package/ios/libs/llama.xcframework/xros-arm64/dSYMs/llama.dSYM/Contents/Resources/DWARF/llama +0 -0
- package/ios/libs/llama.xcframework/xros-arm64/dSYMs/llama.dSYM/Contents/Resources/Relocations/aarch64/llama.yml +5303 -5095
- package/ios/libs/llama.xcframework/xros-arm64/llama.framework/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/xros-arm64/llama.framework/llama +0 -0
- package/ios/libs/llama.xcframework/xros-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/DWARF/llama +0 -0
- package/ios/libs/llama.xcframework/xros-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/Relocations/aarch64/llama.yml +5274 -5066
- package/ios/libs/llama.xcframework/xros-arm64_x86_64-simulator/dSYMs/llama.dSYM/Contents/Resources/Relocations/x86_64/llama.yml +4044 -3919
- package/ios/libs/llama.xcframework/xros-arm64_x86_64-simulator/llama.framework/Headers/llama.h +15 -7
- package/ios/libs/llama.xcframework/xros-arm64_x86_64-simulator/llama.framework/llama +0 -0
- package/package.json +4 -4
There are too many changes on this page to be displayed.
The amount of changes on this page would crash your brower.
You can still verify the content by downloading the package file manually.