llama-cpp-pydist 0.18.0__py3-none-any.whl → 0.20.0__py3-none-any.whl

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.
Files changed (362) hide show
  1. llama_cpp/binaries/{llama-b7376-bin-win-cpu-x64.zip → llama-b7621-bin-win-cpu-x64.zip} +0 -0
  2. llama_cpp_pydist-0.20.0.dist-info/METADATA +4539 -0
  3. {llama_cpp_pydist-0.18.0.dist-info → llama_cpp_pydist-0.20.0.dist-info}/RECORD +358 -318
  4. {llama_cpp_pydist-0.18.0.dist-info → llama_cpp_pydist-0.20.0.dist-info}/WHEEL +1 -1
  5. scripts/generate_changelog.py +10 -0
  6. vendor_llama_cpp_pydist/llama.cpp/.devops/cann.Dockerfile +1 -1
  7. vendor_llama_cpp_pydist/llama.cpp/.devops/cuda-new.Dockerfile +95 -0
  8. vendor_llama_cpp_pydist/llama.cpp/.devops/llama-cli-cann.Dockerfile +3 -2
  9. vendor_llama_cpp_pydist/llama.cpp/.devops/llama-cpp-cuda.srpm.spec +2 -0
  10. vendor_llama_cpp_pydist/llama.cpp/.devops/llama-cpp.srpm.spec +2 -0
  11. vendor_llama_cpp_pydist/llama.cpp/.gemini/settings.json +1 -0
  12. vendor_llama_cpp_pydist/llama.cpp/.github/ISSUE_TEMPLATE/010-bug-compilation.yml +2 -1
  13. vendor_llama_cpp_pydist/llama.cpp/.github/ISSUE_TEMPLATE/011-bug-results.yml +19 -5
  14. vendor_llama_cpp_pydist/llama.cpp/.github/ISSUE_TEMPLATE/019-bug-misc.yml +14 -2
  15. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/build.yml +33 -2
  16. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/docker.yml +25 -13
  17. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/release.yml +12 -48
  18. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/server-webui.yml +225 -0
  19. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/server.yml +0 -264
  20. vendor_llama_cpp_pydist/llama.cpp/.gitignore +1 -0
  21. vendor_llama_cpp_pydist/llama.cpp/AGENTS.md +81 -0
  22. vendor_llama_cpp_pydist/llama.cpp/CLAUDE.md +1 -0
  23. vendor_llama_cpp_pydist/llama.cpp/CODEOWNERS +3 -2
  24. vendor_llama_cpp_pydist/llama.cpp/CONTRIBUTING.md +34 -5
  25. vendor_llama_cpp_pydist/llama.cpp/README.md +4 -2
  26. vendor_llama_cpp_pydist/llama.cpp/SECURITY.md +3 -0
  27. vendor_llama_cpp_pydist/llama.cpp/ci/run.sh +6 -0
  28. vendor_llama_cpp_pydist/llama.cpp/common/CMakeLists.txt +4 -3
  29. vendor_llama_cpp_pydist/llama.cpp/common/arg.cpp +259 -66
  30. vendor_llama_cpp_pydist/llama.cpp/common/arg.h +12 -2
  31. vendor_llama_cpp_pydist/llama.cpp/common/chat-parser.cpp +11 -0
  32. vendor_llama_cpp_pydist/llama.cpp/common/chat-peg-parser.cpp +12 -2
  33. vendor_llama_cpp_pydist/llama.cpp/common/chat.cpp +172 -3
  34. vendor_llama_cpp_pydist/llama.cpp/common/chat.h +1 -0
  35. vendor_llama_cpp_pydist/llama.cpp/common/common.cpp +151 -88
  36. vendor_llama_cpp_pydist/llama.cpp/common/common.h +38 -13
  37. vendor_llama_cpp_pydist/llama.cpp/common/json-schema-to-grammar.cpp +132 -3
  38. vendor_llama_cpp_pydist/llama.cpp/common/json-schema-to-grammar.h +20 -0
  39. vendor_llama_cpp_pydist/llama.cpp/common/peg-parser.cpp +1 -1
  40. vendor_llama_cpp_pydist/llama.cpp/common/preset.cpp +218 -6
  41. vendor_llama_cpp_pydist/llama.cpp/common/preset.h +45 -3
  42. vendor_llama_cpp_pydist/llama.cpp/common/sampling.cpp +67 -54
  43. vendor_llama_cpp_pydist/llama.cpp/common/sampling.h +8 -0
  44. vendor_llama_cpp_pydist/llama.cpp/convert_hf_to_gguf.py +904 -454
  45. vendor_llama_cpp_pydist/llama.cpp/convert_hf_to_gguf_update.py +6 -0
  46. vendor_llama_cpp_pydist/llama.cpp/docs/android/imported-into-android-studio.jpg +0 -0
  47. vendor_llama_cpp_pydist/llama.cpp/docs/android.md +22 -2
  48. vendor_llama_cpp_pydist/llama.cpp/docs/backend/OPENCL.md +1 -1
  49. vendor_llama_cpp_pydist/llama.cpp/docs/backend/SYCL.md +3 -1
  50. vendor_llama_cpp_pydist/llama.cpp/docs/backend/hexagon/CMakeUserPresets.json +2 -0
  51. vendor_llama_cpp_pydist/llama.cpp/docs/backend/hexagon/README.md +5 -5
  52. vendor_llama_cpp_pydist/llama.cpp/docs/backend/hexagon/developer.md +1 -1
  53. vendor_llama_cpp_pydist/llama.cpp/docs/build.md +21 -2
  54. vendor_llama_cpp_pydist/llama.cpp/docs/development/HOWTO-add-model.md +3 -2
  55. vendor_llama_cpp_pydist/llama.cpp/docs/development/parsing.md +2 -2
  56. vendor_llama_cpp_pydist/llama.cpp/docs/docker.md +15 -11
  57. vendor_llama_cpp_pydist/llama.cpp/docs/ops/Metal.csv +360 -322
  58. vendor_llama_cpp_pydist/llama.cpp/docs/ops/SYCL.csv +797 -361
  59. vendor_llama_cpp_pydist/llama.cpp/docs/ops.md +10 -10
  60. vendor_llama_cpp_pydist/llama.cpp/ggml/CMakeLists.txt +18 -1
  61. vendor_llama_cpp_pydist/llama.cpp/ggml/include/ggml-alloc.h +9 -0
  62. vendor_llama_cpp_pydist/llama.cpp/ggml/include/ggml-backend.h +2 -1
  63. vendor_llama_cpp_pydist/llama.cpp/ggml/include/ggml.h +2 -1
  64. vendor_llama_cpp_pydist/llama.cpp/ggml/src/CMakeLists.txt +24 -7
  65. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-alloc.c +56 -12
  66. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-backend.cpp +32 -13
  67. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cann/aclnn_ops.cpp +248 -19
  68. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cann/aclnn_ops.h +3 -0
  69. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cann/common.h +153 -9
  70. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cann/ggml-cann.cpp +21 -172
  71. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/CMakeLists.txt +16 -2
  72. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/arch/arm/repack.cpp +285 -0
  73. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/arch-fallback.h +28 -0
  74. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/ggml-cpu-impl.h +1 -1
  75. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c +51 -6
  76. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/kleidiai/kernels.cpp +86 -25
  77. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/kleidiai/kleidiai.cpp +15 -8
  78. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp +768 -0
  79. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/repack.cpp +286 -0
  80. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/repack.h +8 -0
  81. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h +0 -4
  82. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/vec.cpp +41 -1
  83. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/vec.h +125 -22
  84. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/CMakeLists.txt +42 -1
  85. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/argmax.cu +2 -2
  86. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/argsort.cu +2 -2
  87. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/common.cuh +36 -0
  88. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/cpy.cu +117 -103
  89. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/cumsum.cu +69 -33
  90. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/fattn-common.cuh +3 -1
  91. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/fattn-mma-f16.cuh +2 -2
  92. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/ggml-cuda.cu +67 -31
  93. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mean.cu +3 -0
  94. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mma.cuh +189 -111
  95. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmf.cuh +14 -10
  96. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmq.cu +34 -8
  97. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmq.cuh +204 -42
  98. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/quantize.cu +151 -0
  99. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/quantize.cuh +14 -0
  100. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/ssm-conv.cu +14 -20
  101. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/topk-moe.cu +17 -2
  102. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/topk-moe.cuh +6 -1
  103. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/vendors/cuda.h +4 -0
  104. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/CMakeLists.txt +3 -1
  105. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/ggml-hexagon.cpp +224 -746
  106. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/CMakeLists.txt +2 -1
  107. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/act-ops.c +369 -129
  108. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/htp-dma.c +5 -11
  109. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/htp-dma.h +46 -15
  110. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/htp-msg.h +6 -5
  111. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/hvx-utils.c +154 -47
  112. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/hvx-utils.h +65 -2
  113. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/main.c +4 -1
  114. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/matmul-ops.c +196 -48
  115. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp-utils.h +1 -0
  116. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/op-desc.h +153 -0
  117. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-impl.h +0 -4
  118. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-device.cpp +57 -0
  119. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-device.h +2 -0
  120. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-device.m +12 -0
  121. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-impl.h +20 -0
  122. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-ops.cpp +71 -2
  123. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-ops.h +1 -0
  124. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal.metal +73 -6
  125. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-opencl/ggml-opencl.cpp +134 -13
  126. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-opencl/kernels/cvt.cl +21 -0
  127. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-opencl/kernels/transpose.cl +13 -0
  128. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-rpc/ggml-rpc.cpp +14 -7
  129. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/CMakeLists.txt +42 -1
  130. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/add-id.cpp +77 -0
  131. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/add-id.hpp +8 -0
  132. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/common.hpp +17 -0
  133. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/convert.cpp +15 -0
  134. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/dequantize.hpp +18 -0
  135. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/dpct/helper.hpp +56 -3
  136. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/element_wise.cpp +97 -0
  137. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/element_wise.hpp +4 -0
  138. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/ggml-sycl.cpp +17 -6
  139. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/mmvq.cpp +22 -0
  140. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/pad.cpp +5 -5
  141. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/ssm_conv.cpp +1 -1
  142. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/vecdotq.hpp +58 -0
  143. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/ggml-vulkan.cpp +916 -337
  144. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/count_experts.comp +51 -0
  145. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/cumsum.comp +28 -14
  146. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/cumsum_multipass1.comp +60 -0
  147. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/cumsum_multipass2.comp +66 -0
  148. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/dequant_funcs.glsl +1 -7
  149. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/diag.comp +29 -0
  150. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/flash_attn.comp +3 -0
  151. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/flash_attn_cm1.comp +3 -0
  152. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/generic_head.glsl +2 -0
  153. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/get_rows.comp +2 -2
  154. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/im2col.comp +17 -4
  155. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mat_vec_iq1_s.comp +40 -24
  156. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mat_vec_iq2_xs.comp +42 -24
  157. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mat_vecq.comp +11 -0
  158. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mat_vecq_funcs.glsl +115 -0
  159. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mm.comp +10 -4
  160. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mm_cm2.comp +29 -18
  161. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mm_funcs.glsl +27 -21
  162. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mm_id_funcs.glsl +3 -1
  163. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mmq.comp +10 -4
  164. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_funcs.glsl +11 -4
  165. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_multi.comp +4 -1
  166. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_neox.comp +4 -1
  167. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_norm.comp +4 -1
  168. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_params.glsl +1 -0
  169. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_vision.comp +4 -1
  170. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/soft_max_large1.comp +62 -0
  171. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/soft_max_large2.comp +79 -0
  172. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/soft_max_large3.comp +65 -0
  173. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/soft_max_large_common.glsl +53 -0
  174. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/topk_moe.comp +57 -21
  175. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/types.glsl +312 -6
  176. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/upscale.comp +54 -0
  177. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp +24 -5
  178. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/xielu.comp +35 -0
  179. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml.c +5 -0
  180. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/constants.py +188 -0
  181. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/gguf_writer.py +38 -2
  182. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/tensor_mapping.py +122 -3
  183. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/utility.py +1 -9
  184. vendor_llama_cpp_pydist/llama.cpp/grammars/README.md +6 -3
  185. vendor_llama_cpp_pydist/llama.cpp/include/llama.h +28 -2
  186. vendor_llama_cpp_pydist/llama.cpp/pyrightconfig.json +1 -1
  187. vendor_llama_cpp_pydist/llama.cpp/scripts/compare-logprobs.py +281 -0
  188. vendor_llama_cpp_pydist/llama.cpp/scripts/snapdragon/adb/run-cli.sh +9 -9
  189. vendor_llama_cpp_pydist/llama.cpp/scripts/snapdragon/adb/run-completion.sh +53 -0
  190. vendor_llama_cpp_pydist/llama.cpp/scripts/snapdragon/adb/run-mtmd.sh +65 -0
  191. vendor_llama_cpp_pydist/llama.cpp/scripts/sync-ggml.last +1 -1
  192. vendor_llama_cpp_pydist/llama.cpp/src/CMakeLists.txt +4 -0
  193. vendor_llama_cpp_pydist/llama.cpp/src/llama-adapter.cpp +12 -3
  194. vendor_llama_cpp_pydist/llama.cpp/src/llama-adapter.h +7 -1
  195. vendor_llama_cpp_pydist/llama.cpp/src/llama-arch.cpp +1966 -2248
  196. vendor_llama_cpp_pydist/llama.cpp/src/llama-arch.h +16 -2
  197. vendor_llama_cpp_pydist/llama.cpp/src/llama-chat.cpp +11 -0
  198. vendor_llama_cpp_pydist/llama.cpp/src/llama-chat.h +1 -0
  199. vendor_llama_cpp_pydist/llama.cpp/src/llama-context.cpp +99 -20
  200. vendor_llama_cpp_pydist/llama.cpp/src/llama-context.h +8 -2
  201. vendor_llama_cpp_pydist/llama.cpp/src/llama-graph.cpp +75 -7
  202. vendor_llama_cpp_pydist/llama.cpp/src/llama-graph.h +17 -4
  203. vendor_llama_cpp_pydist/llama.cpp/src/llama-hparams.cpp +3 -9
  204. vendor_llama_cpp_pydist/llama.cpp/src/llama-hparams.h +7 -9
  205. vendor_llama_cpp_pydist/llama.cpp/src/llama-impl.cpp +4 -0
  206. vendor_llama_cpp_pydist/llama.cpp/src/llama-kv-cache.cpp +85 -31
  207. vendor_llama_cpp_pydist/llama.cpp/src/llama-kv-cache.h +20 -3
  208. vendor_llama_cpp_pydist/llama.cpp/src/llama-memory-hybrid.cpp +1 -1
  209. vendor_llama_cpp_pydist/llama.cpp/src/llama-mmap.cpp +130 -28
  210. vendor_llama_cpp_pydist/llama.cpp/src/llama-mmap.h +5 -1
  211. vendor_llama_cpp_pydist/llama.cpp/src/llama-model-loader.cpp +81 -13
  212. vendor_llama_cpp_pydist/llama.cpp/src/llama-model-loader.h +4 -0
  213. vendor_llama_cpp_pydist/llama.cpp/src/llama-model.cpp +380 -68
  214. vendor_llama_cpp_pydist/llama.cpp/src/llama-model.h +14 -2
  215. vendor_llama_cpp_pydist/llama.cpp/src/llama-quant.cpp +1 -1
  216. vendor_llama_cpp_pydist/llama.cpp/src/llama-sampling.cpp +60 -33
  217. vendor_llama_cpp_pydist/llama.cpp/src/llama-sampling.h +3 -0
  218. vendor_llama_cpp_pydist/llama.cpp/src/llama-vocab.cpp +103 -34
  219. vendor_llama_cpp_pydist/llama.cpp/src/llama-vocab.h +2 -0
  220. vendor_llama_cpp_pydist/llama.cpp/src/llama.cpp +680 -1
  221. vendor_llama_cpp_pydist/llama.cpp/src/models/bert.cpp +4 -2
  222. vendor_llama_cpp_pydist/llama.cpp/src/models/cogvlm.cpp +5 -3
  223. vendor_llama_cpp_pydist/llama.cpp/src/models/deepseek2.cpp +1 -1
  224. vendor_llama_cpp_pydist/llama.cpp/src/models/gemma-embedding.cpp +2 -6
  225. vendor_llama_cpp_pydist/llama.cpp/src/models/gemma3.cpp +3 -4
  226. vendor_llama_cpp_pydist/llama.cpp/src/models/gemma3n-iswa.cpp +4 -7
  227. vendor_llama_cpp_pydist/llama.cpp/src/models/glm4-moe.cpp +28 -11
  228. vendor_llama_cpp_pydist/llama.cpp/src/models/glm4.cpp +27 -4
  229. vendor_llama_cpp_pydist/llama.cpp/src/models/llama.cpp +19 -6
  230. vendor_llama_cpp_pydist/llama.cpp/src/models/maincoder.cpp +117 -0
  231. vendor_llama_cpp_pydist/llama.cpp/src/models/mimo2-iswa.cpp +123 -0
  232. vendor_llama_cpp_pydist/llama.cpp/src/models/models.h +23 -5
  233. vendor_llama_cpp_pydist/llama.cpp/src/models/modern-bert.cpp +115 -0
  234. vendor_llama_cpp_pydist/llama.cpp/src/models/nemotron-h.cpp +35 -6
  235. vendor_llama_cpp_pydist/llama.cpp/src/models/plamo3.cpp +128 -0
  236. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen2.cpp +12 -3
  237. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen3next.cpp +81 -266
  238. vendor_llama_cpp_pydist/llama.cpp/src/unicode.cpp +23 -14
  239. vendor_llama_cpp_pydist/llama.cpp/tests/CMakeLists.txt +8 -0
  240. vendor_llama_cpp_pydist/llama.cpp/tests/test-arg-parser.cpp +29 -0
  241. vendor_llama_cpp_pydist/llama.cpp/tests/test-backend-ops.cpp +200 -61
  242. vendor_llama_cpp_pydist/llama.cpp/tests/test-chat.cpp +186 -3
  243. vendor_llama_cpp_pydist/llama.cpp/tests/test-grammar-llguidance.cpp +3 -0
  244. vendor_llama_cpp_pydist/llama.cpp/tests/test-json-schema-to-grammar.cpp +75 -0
  245. vendor_llama_cpp_pydist/llama.cpp/tests/test-state-restore-fragmented.cpp +122 -0
  246. vendor_llama_cpp_pydist/llama.cpp/tests/test-tokenizer-0.cpp +1 -1
  247. vendor_llama_cpp_pydist/llama.cpp/tests/test-tokenizer-1-bpe.cpp +1 -1
  248. vendor_llama_cpp_pydist/llama.cpp/tests/test-tokenizer-1-spm.cpp +1 -1
  249. vendor_llama_cpp_pydist/llama.cpp/tools/CMakeLists.txt +1 -0
  250. vendor_llama_cpp_pydist/llama.cpp/tools/batched-bench/batched-bench.cpp +11 -0
  251. vendor_llama_cpp_pydist/llama.cpp/tools/cli/README.md +187 -0
  252. vendor_llama_cpp_pydist/llama.cpp/tools/cli/cli.cpp +1 -3
  253. vendor_llama_cpp_pydist/llama.cpp/tools/completion/README.md +195 -23
  254. vendor_llama_cpp_pydist/llama.cpp/tools/completion/completion.cpp +11 -17
  255. vendor_llama_cpp_pydist/llama.cpp/tools/cvector-generator/cvector-generator.cpp +3 -3
  256. vendor_llama_cpp_pydist/llama.cpp/tools/fit-params/CMakeLists.txt +8 -0
  257. vendor_llama_cpp_pydist/llama.cpp/tools/fit-params/README.md +55 -0
  258. vendor_llama_cpp_pydist/llama.cpp/tools/fit-params/fit-params.cpp +66 -0
  259. vendor_llama_cpp_pydist/llama.cpp/tools/imatrix/imatrix.cpp +3 -3
  260. vendor_llama_cpp_pydist/llama.cpp/tools/llama-bench/README.md +1 -1
  261. vendor_llama_cpp_pydist/llama.cpp/tools/llama-bench/llama-bench.cpp +18 -1
  262. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/CMakeLists.txt +3 -0
  263. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip-graph.h +7 -1
  264. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip-impl.h +41 -8
  265. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip-model.h +56 -2
  266. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip.cpp +375 -41
  267. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip.h +1 -1
  268. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/models/conformer.cpp +217 -0
  269. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/models/glm4v.cpp +120 -0
  270. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/models/models.h +20 -0
  271. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/models/whisper-enc.cpp +18 -10
  272. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/models/youtuvl.cpp +179 -0
  273. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd-audio.cpp +371 -550
  274. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd-audio.h +19 -28
  275. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd-cli.cpp +22 -5
  276. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd.cpp +34 -14
  277. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd.h +3 -0
  278. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/tests.sh +1 -0
  279. vendor_llama_cpp_pydist/llama.cpp/tools/perplexity/perplexity.cpp +3 -3
  280. vendor_llama_cpp_pydist/llama.cpp/tools/quantize/quantize.cpp +6 -0
  281. vendor_llama_cpp_pydist/llama.cpp/tools/server/CMakeLists.txt +0 -8
  282. vendor_llama_cpp_pydist/llama.cpp/tools/server/README-dev.md +2 -0
  283. vendor_llama_cpp_pydist/llama.cpp/tools/server/README.md +71 -40
  284. vendor_llama_cpp_pydist/llama.cpp/tools/server/public/index.html.gz +0 -0
  285. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-common.cpp +10 -17
  286. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-common.h +2 -3
  287. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-context.cpp +473 -287
  288. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-context.h +52 -15
  289. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-http.cpp +16 -10
  290. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-models.cpp +256 -315
  291. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-models.h +35 -28
  292. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-queue.cpp +78 -21
  293. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-queue.h +48 -10
  294. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-task.cpp +33 -11
  295. vendor_llama_cpp_pydist/llama.cpp/tools/server/server-task.h +28 -35
  296. vendor_llama_cpp_pydist/llama.cpp/tools/server/server.cpp +26 -12
  297. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/unit/test_chat_completion.py +11 -2
  298. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/unit/test_compat_anthropic.py +1 -1
  299. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/unit/test_sleep.py +39 -0
  300. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/utils.py +3 -0
  301. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/README.md +6 -5
  302. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/docs/architecture/high-level-architecture-simplified.md +6 -2
  303. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/docs/architecture/high-level-architecture.md +13 -3
  304. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/package-lock.json +10 -10
  305. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/package.json +3 -4
  306. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/app.d.ts +7 -0
  307. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatAttachments/ChatAttachmentPreview.svelte +1 -1
  308. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatAttachments/ChatAttachmentThumbnailFile.svelte +1 -1
  309. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatForm/ChatForm.svelte +28 -3
  310. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatForm/ChatFormActions/ChatFormActionFileAttachments.svelte +1 -1
  311. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatForm/ChatFormActions/ChatFormActions.svelte +1 -0
  312. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte +53 -8
  313. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte +29 -3
  314. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageEditForm.svelte +391 -0
  315. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageStatistics.svelte +165 -10
  316. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageUser.svelte +26 -48
  317. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessages.svelte +9 -4
  318. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatScreen/ChatScreen.svelte +21 -4
  319. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatScreen/ChatScreenHeader.svelte +6 -1
  320. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettings.svelte +16 -1
  321. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatSettings/ChatSettingsImportExportTab.svelte +68 -2
  322. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatSidebar/ChatSidebar.svelte +5 -1
  323. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/misc/BadgeChatStatistic.svelte +26 -7
  324. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/misc/MarkdownContent.svelte +263 -167
  325. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/misc/SyntaxHighlightedCode.svelte +3 -2
  326. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/models/ModelsSelector.svelte +199 -185
  327. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/server/ServerErrorSplash.svelte +2 -1
  328. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/ui/switch/index.ts +7 -0
  329. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/ui/switch/switch.svelte +29 -0
  330. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/constants/settings-config.ts +9 -0
  331. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/enums/chat.ts +4 -0
  332. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/enums/index.ts +2 -0
  333. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/hooks/use-processing-state.svelte.ts +125 -11
  334. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/markdown/enhance-code-blocks.ts +162 -0
  335. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/markdown/enhance-links.ts +33 -0
  336. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/services/chat.ts +12 -8
  337. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/services/parameter-sync.spec.ts +14 -0
  338. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/services/parameter-sync.ts +75 -13
  339. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/stores/chat.svelte.ts +235 -171
  340. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/stores/conversations.svelte.ts +24 -2
  341. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/stores/server.svelte.ts +4 -0
  342. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/stores/settings.svelte.ts +6 -6
  343. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/types/api.d.ts +4 -0
  344. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/types/settings.d.ts +1 -1
  345. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/api-key-validation.ts +2 -1
  346. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/clipboard.ts +259 -0
  347. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/file-preview.ts +0 -9
  348. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/index.ts +11 -2
  349. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/text.ts +7 -0
  350. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/routes/+layout.svelte +26 -10
  351. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/tests/unit/clipboard.test.ts +423 -0
  352. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/{src/lib/utils → tests/unit}/latex-protection.test.ts +1 -1
  353. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/{src/lib/utils → tests/unit}/model-names.test.ts +1 -1
  354. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/vite.config.ts +2 -2
  355. vendor_llama_cpp_pydist/llama.cpp/tools/tts/tts.cpp +6 -6
  356. vendor_llama_cpp_pydist/llama.cpp/vendor/cpp-httplib/CMakeLists.txt +2 -1
  357. llama_cpp_pydist-0.18.0.dist-info/METADATA +0 -2448
  358. vendor_llama_cpp_pydist/llama.cpp/.github/copilot-instructions.md +0 -262
  359. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/copy.ts +0 -71
  360. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/tests/server/demo.spec.ts +0 -7
  361. {llama_cpp_pydist-0.18.0.dist-info/licenses → llama_cpp_pydist-0.20.0.dist-info}/LICENSE +0 -0
  362. {llama_cpp_pydist-0.18.0.dist-info → llama_cpp_pydist-0.20.0.dist-info}/top_level.txt +0 -0