llama-cpp-pydist 0.13.0__py3-none-any.whl → 0.15.0__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 (279) hide show
  1. llama_cpp/binaries/{llama-b6916-bin-win-cpu-x64.zip → llama-b7058-bin-win-cpu-x64.zip} +0 -0
  2. {llama_cpp_pydist-0.13.0.dist-info → llama_cpp_pydist-0.15.0.dist-info}/METADATA +118 -1
  3. {llama_cpp_pydist-0.13.0.dist-info → llama_cpp_pydist-0.15.0.dist-info}/RECORD +277 -252
  4. vendor_llama_cpp_pydist/llama.cpp/.devops/cann.Dockerfile +1 -1
  5. vendor_llama_cpp_pydist/llama.cpp/.devops/cpu.Dockerfile +1 -1
  6. vendor_llama_cpp_pydist/llama.cpp/.devops/cuda.Dockerfile +1 -1
  7. vendor_llama_cpp_pydist/llama.cpp/.devops/intel.Dockerfile +1 -1
  8. vendor_llama_cpp_pydist/llama.cpp/.devops/musa.Dockerfile +1 -1
  9. vendor_llama_cpp_pydist/llama.cpp/.devops/nix/package.nix +2 -0
  10. vendor_llama_cpp_pydist/llama.cpp/.devops/rocm.Dockerfile +1 -1
  11. vendor_llama_cpp_pydist/llama.cpp/.devops/s390x.Dockerfile +4 -1
  12. vendor_llama_cpp_pydist/llama.cpp/.devops/vulkan.Dockerfile +5 -21
  13. vendor_llama_cpp_pydist/llama.cpp/.editorconfig +8 -0
  14. vendor_llama_cpp_pydist/llama.cpp/.github/copilot-instructions.md +1 -1
  15. vendor_llama_cpp_pydist/llama.cpp/.github/labeler.yml +4 -0
  16. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/build-linux-cross.yml +37 -37
  17. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/build.yml +57 -8
  18. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/check-vendor.yml +52 -0
  19. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/release.yml +2 -2
  20. vendor_llama_cpp_pydist/llama.cpp/.github/workflows/server.yml +1 -1
  21. vendor_llama_cpp_pydist/llama.cpp/CMakeLists.txt +4 -0
  22. vendor_llama_cpp_pydist/llama.cpp/README.md +4 -4
  23. vendor_llama_cpp_pydist/llama.cpp/benches/dgx-spark/aime25_openai__gpt-oss-120b-high_temp1.0_20251109_094547.html +35144 -0
  24. vendor_llama_cpp_pydist/llama.cpp/benches/dgx-spark/aime25_openai__gpt-oss-120b-high_temp1.0_20251109_094547.json +6 -0
  25. vendor_llama_cpp_pydist/llama.cpp/benches/dgx-spark/aime25_openai__gpt-oss-120b-high_temp1.0_20251109_094547_allresults.json +2896 -0
  26. vendor_llama_cpp_pydist/llama.cpp/benches/dgx-spark/dgx-spark.md +264 -0
  27. vendor_llama_cpp_pydist/llama.cpp/benches/dgx-spark/run-aime-120b-t8-x8-high.log +11 -0
  28. vendor_llama_cpp_pydist/llama.cpp/build-xcframework.sh +4 -0
  29. vendor_llama_cpp_pydist/llama.cpp/ci/run.sh +6 -1
  30. vendor_llama_cpp_pydist/llama.cpp/common/CMakeLists.txt +8 -37
  31. vendor_llama_cpp_pydist/llama.cpp/common/arg.cpp +42 -997
  32. vendor_llama_cpp_pydist/llama.cpp/common/arg.h +2 -2
  33. vendor_llama_cpp_pydist/llama.cpp/common/chat.cpp +17 -2
  34. vendor_llama_cpp_pydist/llama.cpp/common/common.cpp +33 -0
  35. vendor_llama_cpp_pydist/llama.cpp/common/common.h +15 -1
  36. vendor_llama_cpp_pydist/llama.cpp/common/download.cpp +1072 -0
  37. vendor_llama_cpp_pydist/llama.cpp/common/download.h +55 -0
  38. vendor_llama_cpp_pydist/llama.cpp/convert_hf_to_gguf.py +324 -16
  39. vendor_llama_cpp_pydist/llama.cpp/convert_hf_to_gguf_update.py +1 -0
  40. vendor_llama_cpp_pydist/llama.cpp/docs/backend/CANN.md +6 -1
  41. vendor_llama_cpp_pydist/llama.cpp/docs/backend/OPENCL.md +25 -3
  42. vendor_llama_cpp_pydist/llama.cpp/docs/build.md +42 -0
  43. vendor_llama_cpp_pydist/llama.cpp/docs/docker.md +3 -3
  44. vendor_llama_cpp_pydist/llama.cpp/docs/ops/CPU.csv +16067 -5133
  45. vendor_llama_cpp_pydist/llama.cpp/docs/ops/CUDA.csv +16075 -5125
  46. vendor_llama_cpp_pydist/llama.cpp/docs/ops/SYCL.csv +2404 -2289
  47. vendor_llama_cpp_pydist/llama.cpp/docs/ops.md +29 -23
  48. vendor_llama_cpp_pydist/llama.cpp/ggml/CMakeLists.txt +1 -1
  49. vendor_llama_cpp_pydist/llama.cpp/ggml/include/ggml.h +72 -0
  50. vendor_llama_cpp_pydist/llama.cpp/ggml/src/CMakeLists.txt +22 -3
  51. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-backend.cpp +0 -2
  52. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cann/aclnn_ops.cpp +115 -0
  53. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cann/aclnn_ops.h +62 -0
  54. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cann/ggml-cann.cpp +8 -0
  55. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/CMakeLists.txt +44 -14
  56. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c +428 -26
  57. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/arch/loongarch/quants.c +4 -5
  58. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/arch/riscv/quants.c +108 -49
  59. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/arch/s390/cpu-feats.cpp +50 -0
  60. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/ggml-cpu-impl.h +3 -1
  61. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c +50 -16
  62. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/kleidiai/kernels.cpp +283 -0
  63. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/kleidiai/kernels.h +1 -0
  64. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/kleidiai/kleidiai.cpp +235 -34
  65. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/ops.cpp +346 -325
  66. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/ops.h +4 -4
  67. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/repack.cpp +111 -33
  68. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h +25 -25
  69. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/unary-ops.cpp +16 -0
  70. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/unary-ops.h +2 -0
  71. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/vec.cpp +17 -0
  72. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cpu/vec.h +10 -0
  73. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/CMakeLists.txt +1 -0
  74. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/common.cuh +6 -0
  75. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/cpy.cu +94 -7
  76. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/fattn-tile.cu +4 -0
  77. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/fattn-tile.cuh +29 -2
  78. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/fattn.cu +3 -2
  79. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/ggml-cuda.cu +104 -31
  80. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmf.cu +15 -3
  81. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmf.cuh +1 -1
  82. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmq.cuh +1 -1
  83. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmvf.cu +14 -1
  84. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/mmvf.cuh +1 -1
  85. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/rope.cu +162 -60
  86. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/rope.cuh +2 -0
  87. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/template-instances/fattn-tile-instance-dkq72-dv72.cu +5 -0
  88. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/template-instances/generate_cu_files.py +3 -1
  89. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/unary.cu +48 -0
  90. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/unary.cuh +12 -0
  91. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/upscale.cu +87 -6
  92. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/ggml-hexagon.cpp +34 -31
  93. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/binary-ops.c +46 -30
  94. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/htp-msg.h +4 -4
  95. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/htp-ops.h +11 -0
  96. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp/ops-utils.h +50 -17
  97. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-hexagon/htp-utils.h +1 -0
  98. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-impl.h +1 -1
  99. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-context.m +5 -6
  100. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-device.cpp +52 -0
  101. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-device.h +6 -1
  102. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-device.m +213 -8
  103. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-impl.h +50 -2
  104. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-ops.cpp +166 -19
  105. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal-ops.h +1 -0
  106. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal.cpp +4 -0
  107. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-metal/ggml-metal.metal +655 -152
  108. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-opencl/ggml-opencl.cpp +47 -4
  109. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-opencl/kernels/rope.cl +50 -24
  110. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-opencl/kernels/set_rows.cl +35 -16
  111. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/concat.cpp +55 -44
  112. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/ggml-sycl.cpp +2 -5
  113. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-sycl/repeat_back.cpp +45 -25
  114. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/ggml-vulkan.cpp +1228 -1421
  115. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/conv2d_mm.comp +39 -36
  116. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/generic_binary_head.glsl +8 -0
  117. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mat_vec_base.glsl +19 -0
  118. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mm.comp +31 -2
  119. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mmq.comp +4 -2
  120. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/mul_mmq_funcs.glsl +22 -11
  121. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/quantize_q8_1.comp +1 -1
  122. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rms_norm.comp +43 -1
  123. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_funcs.glsl +227 -0
  124. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_head.glsl +9 -47
  125. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_multi.comp +4 -63
  126. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_neox.comp +4 -41
  127. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_norm.comp +4 -41
  128. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_params.glsl +27 -0
  129. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/rope_vision.comp +4 -40
  130. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/upscale.comp +37 -0
  131. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp +55 -34
  132. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/ggml-webgpu.cpp +377 -35
  133. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/embed_wgsl.py +6 -3
  134. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/mul_mat.tmpl.wgsl +5 -5
  135. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_decls.tmpl +97 -0
  136. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_reg_tile.tmpl.wgsl +247 -0
  137. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_subgroup_matrix.tmpl.wgsl +302 -0
  138. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_vec.tmpl.wgsl +267 -0
  139. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-webgpu/wgsl-shaders/{set_rows.wgsl → set_rows.tmpl.wgsl} +38 -7
  140. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml.c +154 -5
  141. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/constants.py +52 -0
  142. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/lazy.py +8 -3
  143. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/tensor_mapping.py +10 -1
  144. vendor_llama_cpp_pydist/llama.cpp/gguf-py/gguf/utility.py +80 -0
  145. vendor_llama_cpp_pydist/llama.cpp/include/llama.h +6 -3
  146. vendor_llama_cpp_pydist/llama.cpp/scripts/bench-models.sh +74 -0
  147. vendor_llama_cpp_pydist/llama.cpp/scripts/sync-ggml.last +1 -1
  148. vendor_llama_cpp_pydist/llama.cpp/scripts/sync_vendor.py +21 -2
  149. vendor_llama_cpp_pydist/llama.cpp/src/CMakeLists.txt +7 -0
  150. vendor_llama_cpp_pydist/llama.cpp/src/llama-arch.cpp +50 -0
  151. vendor_llama_cpp_pydist/llama.cpp/src/llama-arch.h +3 -0
  152. vendor_llama_cpp_pydist/llama.cpp/src/llama-chat.cpp +32 -0
  153. vendor_llama_cpp_pydist/llama.cpp/src/llama-chat.h +1 -0
  154. vendor_llama_cpp_pydist/llama.cpp/src/llama-context.cpp +36 -13
  155. vendor_llama_cpp_pydist/llama.cpp/src/llama-context.h +5 -5
  156. vendor_llama_cpp_pydist/llama.cpp/src/llama-cparams.h +1 -0
  157. vendor_llama_cpp_pydist/llama.cpp/src/llama-graph.cpp +4 -3
  158. vendor_llama_cpp_pydist/llama.cpp/src/llama-hparams.cpp +10 -0
  159. vendor_llama_cpp_pydist/llama.cpp/src/llama-hparams.h +3 -0
  160. vendor_llama_cpp_pydist/llama.cpp/src/llama-kv-cache-iswa.cpp +3 -1
  161. vendor_llama_cpp_pydist/llama.cpp/src/llama-memory-recurrent.cpp +4 -3
  162. vendor_llama_cpp_pydist/llama.cpp/src/llama-model.cpp +171 -26
  163. vendor_llama_cpp_pydist/llama.cpp/src/llama-model.h +2 -0
  164. vendor_llama_cpp_pydist/llama.cpp/src/llama-sampling.cpp +10 -5
  165. vendor_llama_cpp_pydist/llama.cpp/src/llama-vocab.cpp +16 -1
  166. vendor_llama_cpp_pydist/llama.cpp/src/llama-vocab.h +1 -0
  167. vendor_llama_cpp_pydist/llama.cpp/src/models/afmoe.cpp +187 -0
  168. vendor_llama_cpp_pydist/llama.cpp/src/models/ernie4-5.cpp +4 -5
  169. vendor_llama_cpp_pydist/llama.cpp/src/models/gemma2-iswa.cpp +113 -113
  170. vendor_llama_cpp_pydist/llama.cpp/src/models/gemma3-iswa.cpp +116 -116
  171. vendor_llama_cpp_pydist/llama.cpp/src/models/glm4-moe.cpp +139 -139
  172. vendor_llama_cpp_pydist/llama.cpp/src/models/grok.cpp +133 -134
  173. vendor_llama_cpp_pydist/llama.cpp/src/models/hunyuan-dense.cpp +113 -113
  174. vendor_llama_cpp_pydist/llama.cpp/src/models/hunyuan-moe.cpp +135 -135
  175. vendor_llama_cpp_pydist/llama.cpp/src/models/internlm2.cpp +97 -98
  176. vendor_llama_cpp_pydist/llama.cpp/src/models/jais.cpp +70 -70
  177. vendor_llama_cpp_pydist/llama.cpp/src/models/jamba.cpp +97 -98
  178. vendor_llama_cpp_pydist/llama.cpp/src/models/llada-moe.cpp +113 -114
  179. vendor_llama_cpp_pydist/llama.cpp/src/models/llada.cpp +71 -73
  180. vendor_llama_cpp_pydist/llama.cpp/src/models/llama.cpp +128 -129
  181. vendor_llama_cpp_pydist/llama.cpp/src/models/minicpm3.cpp +177 -178
  182. vendor_llama_cpp_pydist/llama.cpp/src/models/models.h +8 -0
  183. vendor_llama_cpp_pydist/llama.cpp/src/models/nemotron.cpp +99 -99
  184. vendor_llama_cpp_pydist/llama.cpp/src/models/neo-bert.cpp +97 -97
  185. vendor_llama_cpp_pydist/llama.cpp/src/models/olmo.cpp +98 -98
  186. vendor_llama_cpp_pydist/llama.cpp/src/models/olmo2.cpp +122 -123
  187. vendor_llama_cpp_pydist/llama.cpp/src/models/olmoe.cpp +107 -107
  188. vendor_llama_cpp_pydist/llama.cpp/src/models/openai-moe-iswa.cpp +106 -105
  189. vendor_llama_cpp_pydist/llama.cpp/src/models/openelm.cpp +105 -105
  190. vendor_llama_cpp_pydist/llama.cpp/src/models/orion.cpp +113 -113
  191. vendor_llama_cpp_pydist/llama.cpp/src/models/pangu-embedded.cpp +121 -0
  192. vendor_llama_cpp_pydist/llama.cpp/src/models/phi3.cpp +128 -129
  193. vendor_llama_cpp_pydist/llama.cpp/src/models/plamo.cpp +100 -100
  194. vendor_llama_cpp_pydist/llama.cpp/src/models/plm.cpp +154 -154
  195. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen2.cpp +108 -109
  196. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen2moe.cpp +127 -127
  197. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen2vl.cpp +107 -107
  198. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen3.cpp +107 -107
  199. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen3moe.cpp +115 -115
  200. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen3vl-moe.cpp +1 -2
  201. vendor_llama_cpp_pydist/llama.cpp/src/models/qwen3vl.cpp +1 -4
  202. vendor_llama_cpp_pydist/llama.cpp/src/models/refact.cpp +76 -76
  203. vendor_llama_cpp_pydist/llama.cpp/src/models/seed-oss.cpp +100 -100
  204. vendor_llama_cpp_pydist/llama.cpp/src/models/smallthinker.cpp +92 -92
  205. vendor_llama_cpp_pydist/llama.cpp/src/models/smollm3.cpp +100 -100
  206. vendor_llama_cpp_pydist/llama.cpp/src/models/starcoder.cpp +81 -81
  207. vendor_llama_cpp_pydist/llama.cpp/src/models/starcoder2.cpp +93 -93
  208. vendor_llama_cpp_pydist/llama.cpp/src/models/t5-dec.cpp +135 -135
  209. vendor_llama_cpp_pydist/llama.cpp/src/models/t5-enc.cpp +77 -77
  210. vendor_llama_cpp_pydist/llama.cpp/src/models/wavtokenizer-dec.cpp +106 -106
  211. vendor_llama_cpp_pydist/llama.cpp/src/models/xverse.cpp +89 -89
  212. vendor_llama_cpp_pydist/llama.cpp/src/unicode.cpp +77 -0
  213. vendor_llama_cpp_pydist/llama.cpp/tests/test-backend-ops.cpp +538 -148
  214. vendor_llama_cpp_pydist/llama.cpp/tests/test-rope.cpp +6 -5
  215. vendor_llama_cpp_pydist/llama.cpp/tests/test-thread-safety.cpp +8 -1
  216. vendor_llama_cpp_pydist/llama.cpp/tools/batched-bench/batched-bench.cpp +31 -13
  217. vendor_llama_cpp_pydist/llama.cpp/tools/llama-bench/llama-bench.cpp +40 -7
  218. vendor_llama_cpp_pydist/llama.cpp/tools/main/main.cpp +5 -1
  219. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/CMakeLists.txt +5 -0
  220. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip-impl.h +2 -0
  221. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip.cpp +314 -175
  222. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/clip.h +10 -0
  223. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd-cli.cpp +7 -4
  224. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd-helper.cpp +1 -1
  225. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd.cpp +30 -15
  226. vendor_llama_cpp_pydist/llama.cpp/tools/mtmd/mtmd.h +5 -0
  227. vendor_llama_cpp_pydist/llama.cpp/tools/rpc/CMakeLists.txt +4 -0
  228. vendor_llama_cpp_pydist/llama.cpp/tools/server/CMakeLists.txt +5 -1
  229. vendor_llama_cpp_pydist/llama.cpp/tools/server/README.md +2 -2
  230. vendor_llama_cpp_pydist/llama.cpp/tools/server/public/index.html.gz +0 -0
  231. vendor_llama_cpp_pydist/llama.cpp/tools/server/server.cpp +387 -284
  232. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/unit/test_chat_completion.py +4 -4
  233. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/unit/test_completion.py +73 -0
  234. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/unit/test_infill.py +2 -2
  235. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/unit/test_speculative.py +6 -6
  236. vendor_llama_cpp_pydist/llama.cpp/tools/server/tests/utils.py +3 -0
  237. vendor_llama_cpp_pydist/llama.cpp/tools/server/utils.hpp +22 -15
  238. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/.storybook/preview.ts +8 -0
  239. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/.storybook/vitest.setup.ts +2 -1
  240. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/package-lock.json +541 -305
  241. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/package.json +14 -13
  242. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatAttachments/ChatAttachmentFilePreview.svelte +19 -29
  243. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatAttachments/ChatAttachmentImagePreview.svelte +5 -14
  244. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatAttachments/ChatAttachmentPreviewDialog.svelte +10 -1
  245. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatAttachments/ChatAttachmentsList.svelte +143 -60
  246. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatAttachments/ChatAttachmentsViewAllDialog.svelte +203 -0
  247. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatForm/ChatForm.svelte +7 -1
  248. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessageAssistant.svelte +2 -2
  249. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/chat/ChatScreen/ChatScreen.svelte +9 -11
  250. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/index.ts +3 -0
  251. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/misc/MarkdownContent.svelte +3 -12
  252. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/components/app/misc/RemoveButton.svelte +26 -0
  253. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/constants/latex-protection.ts +35 -0
  254. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/services/chat.ts +16 -4
  255. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/stores/chat.svelte.ts +52 -2
  256. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/stores/server.svelte.ts +113 -64
  257. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/types/api.d.ts +1 -0
  258. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/types/chat.d.ts +23 -0
  259. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/types/settings.d.ts +1 -0
  260. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/latex-protection.test.ts +355 -0
  261. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/lib/utils/latex-protection.ts +267 -0
  262. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/routes/+layout.svelte +2 -2
  263. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/stories/ChatForm.stories.svelte +1 -1
  264. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/stories/ChatSidebar.stories.svelte +1 -1
  265. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/stories/MarkdownContent.stories.svelte +25 -26
  266. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/stories/fixtures/math-formulas.ts +65 -0
  267. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/src/styles/katex-custom.scss +13 -0
  268. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/svelte.config.js +3 -0
  269. vendor_llama_cpp_pydist/llama.cpp/tools/server/webui/vite.config.ts +28 -3
  270. vendor_llama_cpp_pydist/llama.cpp/vendor/cpp-httplib/CMakeLists.txt +60 -0
  271. vendor_llama_cpp_pydist/llama.cpp/vendor/cpp-httplib/httplib.cpp +9339 -0
  272. vendor_llama_cpp_pydist/llama.cpp/vendor/cpp-httplib/httplib.h +0 -9336
  273. vendor_llama_cpp_pydist/llama.cpp/vendor/minja/chat-template.hpp +9 -2
  274. vendor_llama_cpp_pydist/llama.cpp/vendor/minja/minja.hpp +5 -7
  275. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/moe-expert-reduce.cu +0 -168
  276. vendor_llama_cpp_pydist/llama.cpp/ggml/src/ggml-cuda/moe-expert-reduce.cuh +0 -11
  277. {llama_cpp_pydist-0.13.0.dist-info → llama_cpp_pydist-0.15.0.dist-info}/WHEEL +0 -0
  278. {llama_cpp_pydist-0.13.0.dist-info → llama_cpp_pydist-0.15.0.dist-info}/licenses/LICENSE +0 -0
  279. {llama_cpp_pydist-0.13.0.dist-info → llama_cpp_pydist-0.15.0.dist-info}/top_level.txt +0 -0
{llama_cpp_pydist-0.13.0.dist-info → llama_cpp_pydist-0.15.0.dist-info}/METADATA RENAMED
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: llama_cpp_pydist
3
- Version: 0.13.0
3
+ Version: 0.15.0
4
4
  Summary: A Python package for Llama CPP.
5
5
  Home-page: https://github.com/shamitv/llama_cpp
6
6
  Author: Shamit Verma
@@ -146,6 +146,123 @@ For instructions on how to build the package from source, update the `llama.cpp`
146
146
 
147
147
  # Changelog
148
148
 
149
+ ## 2025-11-14: Update to llama.cpp b7058
150
+
151
+ - b6959 (b6959) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6959
152
+ - b6960 (b6960) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6960
153
+ - b6961 (b6961) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6961
154
+ - b6962 (b6962) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6962
155
+ - b6963 (b6963) – 2025-11-06 – https://github.com/ggml-org/llama.cpp/releases/tag/b6963
156
+ - b6965 (b6965) – 2025-11-06 – https://github.com/ggml-org/llama.cpp/releases/tag/b6965
157
+ - b6966 (b6966) – 2025-11-06 – https://github.com/ggml-org/llama.cpp/releases/tag/b6966
158
+ - b6967 (b6967) – 2025-11-06 – https://github.com/ggml-org/llama.cpp/releases/tag/b6967
159
+ - b6968 (b6968) – 2025-11-06 – https://github.com/ggml-org/llama.cpp/releases/tag/b6968
160
+ - b6969 (b6969) – 2025-11-06 – https://github.com/ggml-org/llama.cpp/releases/tag/b6969
161
+ - b6970 (b6970) – 2025-11-06 – https://github.com/ggml-org/llama.cpp/releases/tag/b6970
162
+ - b6971 (b6971) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6971
163
+ - b6972 (b6972) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6972
164
+ - b6973 (b6973) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6973
165
+ - b6974 (b6974) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6974
166
+ - b6975 (b6975) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6975
167
+ - b6976 (b6976) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6976
168
+ - b6977 (b6977) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6977
169
+ - b6978 (b6978) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6978
170
+ - b6979 (b6979) – 2025-11-07 – https://github.com/ggml-org/llama.cpp/releases/tag/b6979
171
+ - b6980 (b6980) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6980
172
+ - b6981 (b6981) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6981
173
+ - b6982 (b6982) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6982
174
+ - b6983 (b6983) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6983
175
+ - b6984 (b6984) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6984
176
+ - b6985 (b6985) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6985
177
+ - b6986 (b6986) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6986
178
+ - b6987 (b6987) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6987
179
+ - b6988 (b6988) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6988
180
+ - b6989 (b6989) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6989
181
+ - b6990 (b6990) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6990
182
+ - b6992 (b6992) – 2025-11-08 – https://github.com/ggml-org/llama.cpp/releases/tag/b6992
183
+ - b6993 (b6993) – 2025-11-09 – https://github.com/ggml-org/llama.cpp/releases/tag/b6993
184
+ - b6994 (b6994) – 2025-11-09 – https://github.com/ggml-org/llama.cpp/releases/tag/b6994
185
+ - b6995 (b6995) – 2025-11-09 – https://github.com/ggml-org/llama.cpp/releases/tag/b6995
186
+ - b6996 (b6996) – 2025-11-09 – https://github.com/ggml-org/llama.cpp/releases/tag/b6996
187
+ - b6999 (b6999) – 2025-11-09 – https://github.com/ggml-org/llama.cpp/releases/tag/b6999
188
+ - b7002 (b7002) – 2025-11-09 – https://github.com/ggml-org/llama.cpp/releases/tag/b7002
189
+ - b7003 (b7003) – 2025-11-09 – https://github.com/ggml-org/llama.cpp/releases/tag/b7003
190
+ - b7005 (b7005) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7005
191
+ - b7007 (b7007) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7007
192
+ - b7008 (b7008) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7008
193
+ - b7009 (b7009) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7009
194
+ - b7010 (b7010) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7010
195
+ - b7011 (b7011) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7011
196
+ - b7012 (b7012) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7012
197
+ - b7013 (b7013) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7013
198
+ - b7014 (b7014) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7014
199
+ - b7015 (b7015) – 2025-11-10 – https://github.com/ggml-org/llama.cpp/releases/tag/b7015
200
+ - b7016 (b7016) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7016
201
+ - b7017 (b7017) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7017
202
+ - b7018 (b7018) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7018
203
+ - b7020 (b7020) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7020
204
+ - b7021 (b7021) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7021
205
+ - b7022 (b7022) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7022
206
+ - b7023 (b7023) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7023
207
+ - b7024 (b7024) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7024
208
+ - b7025 (b7025) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7025
209
+ - b7027 (b7027) – 2025-11-11 – https://github.com/ggml-org/llama.cpp/releases/tag/b7027
210
+ - b7028 (b7028) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7028
211
+ - b7030 (b7030) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7030
212
+ - b7031 (b7031) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7031
213
+ - b7032 (b7032) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7032
214
+ - b7033 (b7033) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7033
215
+ - b7034 (b7034) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7034
216
+ - b7035 (b7035) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7035
217
+ - b7037 (b7037) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7037
218
+ - b7039 (b7039) – 2025-11-12 – https://github.com/ggml-org/llama.cpp/releases/tag/b7039
219
+ - b7041 (b7041) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7041
220
+ - b7042 (b7042) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7042
221
+ - b7044 (b7044) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7044
222
+ - b7045 (b7045) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7045
223
+ - b7046 (b7046) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7046
224
+ - b7047 (b7047) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7047
225
+ - b7048 (b7048) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7048
226
+ - b7049 (b7049) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7049
227
+ - b7050 (b7050) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7050
228
+ - b7051 (b7051) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7051
229
+ - b7052 (b7052) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7052
230
+ - b7053 (b7053) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7053
231
+ - b7054 (b7054) – 2025-11-13 – https://github.com/ggml-org/llama.cpp/releases/tag/b7054
232
+ - b7057 (b7057) – 2025-11-14 – https://github.com/ggml-org/llama.cpp/releases/tag/b7057
233
+ - b7058 (b7058) – 2025-11-14 – https://github.com/ggml-org/llama.cpp/releases/tag/b7058
234
+
235
+
236
+ ## 2025-11-05: Update to llama.cpp b6957
237
+
238
+ - b6919 (b6919) – 2025-11-01 – https://github.com/ggml-org/llama.cpp/releases/tag/b6919
239
+ - b6920 (b6920) – 2025-11-02 – https://github.com/ggml-org/llama.cpp/releases/tag/b6920
240
+ - b6922 (b6922) – 2025-11-02 – https://github.com/ggml-org/llama.cpp/releases/tag/b6922
241
+ - b6923 (b6923) – 2025-11-02 – https://github.com/ggml-org/llama.cpp/releases/tag/b6923
242
+ - b6924 (b6924) – 2025-11-02 – https://github.com/ggml-org/llama.cpp/releases/tag/b6924
243
+ - b6927 (b6927) – 2025-11-02 – https://github.com/ggml-org/llama.cpp/releases/tag/b6927
244
+ - b6929 (b6929) – 2025-11-02 – https://github.com/ggml-org/llama.cpp/releases/tag/b6929
245
+ - b6931 (b6931) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6931
246
+ - b6932 (b6932) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6932
247
+ - b6933 (b6933) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6933
248
+ - b6934 (b6934) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6934
249
+ - b6935 (b6935) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6935
250
+ - b6936 (b6936) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6936
251
+ - b6937 (b6937) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6937
252
+ - b6940 (b6940) – 2025-11-03 – https://github.com/ggml-org/llama.cpp/releases/tag/b6940
253
+ - b6941 (b6941) – 2025-11-04 – https://github.com/ggml-org/llama.cpp/releases/tag/b6941
254
+ - b6942 (b6942) – 2025-11-04 – https://github.com/ggml-org/llama.cpp/releases/tag/b6942
255
+ - b6943 (b6943) – 2025-11-04 – https://github.com/ggml-org/llama.cpp/releases/tag/b6943
256
+ - b6945 (b6945) – 2025-11-04 – https://github.com/ggml-org/llama.cpp/releases/tag/b6945
257
+ - b6947 (b6947) – 2025-11-04 – https://github.com/ggml-org/llama.cpp/releases/tag/b6947
258
+ - b6948 (b6948) – 2025-11-04 – https://github.com/ggml-org/llama.cpp/releases/tag/b6948
259
+ - b6949 (b6949) – 2025-11-04 – https://github.com/ggml-org/llama.cpp/releases/tag/b6949
260
+ - b6953 (b6953) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6953
261
+ - b6954 (b6954) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6954
262
+ - b6955 (b6955) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6955
263
+ - b6957 (b6957) – 2025-11-05 – https://github.com/ggml-org/llama.cpp/releases/tag/b6957
264
+
265
+
149
266
  ## 2025-11-01: Update to llama.cpp b6916
150
267
 
151
268
  - b6904 (b6904) – 2025-10-31 – https://github.com/ggml-org/llama.cpp/releases/tag/b6904