llama-cpp-python 0.2.84__tar.gz → 0.2.85__tar.gz

llama_cpp_python-0.2.85/.git/FETCH_HEAD

@@ -0,0 +1 @@
+f7b9e6d42981bb75106699e0712a6378c39ef92c		'f7b9e6d42981bb75106699e0712a6378c39ef92c' of https://github.com/abetlen/llama-cpp-python

llama_cpp_python-0.2.85/.git/HEAD

@@ -0,0 +1 @@
+f7b9e6d42981bb75106699e0712a6378c39ef92c

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/.git/config

@@ -9,7 +9,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX1RBM0k4NmYwRDdyeUJydTZJajU3akNZWEhMSEtIdDRPR1l5Nw==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX0dGSTFocVdiMWlLV1RxdUkyTUxyU1FNR0RmS2dDQzNiUTI0eA==
 [submodule "vendor/llama.cpp"]
 	active = true
 	url = https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.2.85/.git/logs/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 f7b9e6d42981bb75106699e0712a6378c39ef92c runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443110 +0000	checkout: moving from master to refs/tags/v0.2.85

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/HEAD

@@ -0,0 +1 @@
+398ede5efeb07b9adf9fbda7ea63f630d476a792

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/.git/modules/vendor/llama.cpp/config

@@ -16,7 +16,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX1RBM0k4NmYwRDdyeUJydTZJajU3akNZWEhMSEtIdDRPR1l5Nw==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX0dGSTFocVdiMWlLV1RxdUkyTUxyU1FNR0RmS2dDQzNiUTI0eA==
 [url "https://github.com/"]
 	insteadOf = git@github.com:
 	insteadOf = org-6826477@github.com:

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/logs/HEAD

@@ -0,0 +1,2 @@
+0000000000000000000000000000000000000000 398ede5efeb07b9adf9fbda7ea63f630d476a792 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443111 +0000	clone: from https://github.com/ggerganov/llama.cpp.git
+398ede5efeb07b9adf9fbda7ea63f630d476a792 398ede5efeb07b9adf9fbda7ea63f630d476a792 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443111 +0000	checkout: moving from master to 398ede5efeb07b9adf9fbda7ea63f630d476a792

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/logs/refs/heads/master

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 398ede5efeb07b9adf9fbda7ea63f630d476a792 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443111 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/logs/refs/remotes/origin/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 398ede5efeb07b9adf9fbda7ea63f630d476a792 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443111 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/.git/modules/vendor/llama.cpp/modules/kompute/config

@@ -13,7 +13,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX1RBM0k4NmYwRDdyeUJydTZJajU3akNZWEhMSEtIdDRPR1l5Nw==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX0dGSTFocVdiMWlLV1RxdUkyTUxyU1FNR0RmS2dDQzNiUTI0eA==
 [url "https://github.com/"]
 	insteadOf = git@github.com:
 	insteadOf = org-6826477@github.com:

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/modules/kompute/logs/HEAD

@@ -0,0 +1,2 @@
+0000000000000000000000000000000000000000 f592b5bca3cbc169feb194218a086b18d618cca4 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443112 +0000	clone: from https://github.com/nomic-ai/kompute.git
+f592b5bca3cbc169feb194218a086b18d618cca4 4565194ed7c32d1d2efa32ceab4d3c6cae006306 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443112 +0000	checkout: moving from master to 4565194ed7c32d1d2efa32ceab4d3c6cae006306

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/heads/master

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 f592b5bca3cbc169feb194218a086b18d618cca4 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443112 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/remotes/origin/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 f592b5bca3cbc169feb194218a086b18d618cca4 runner <runner@fv-az1780-24.agtqprkliaoehfcdqtdennvexe.cx.internal.cloudapp.net> 1722443112 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/packed-refs

@@ -0,0 +1,2 @@
+# pack-refs with: peeled fully-peeled sorted 
+398ede5efeb07b9adf9fbda7ea63f630d476a792 refs/remotes/origin/master

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/refs/heads/master

@@ -0,0 +1 @@
+398ede5efeb07b9adf9fbda7ea63f630d476a792

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/refs/tags/b3496

@@ -0,0 +1 @@
+398ede5efeb07b9adf9fbda7ea63f630d476a792

llama_cpp_python-0.2.85/.git/modules/vendor/llama.cpp/shallow

@@ -0,0 +1 @@
+398ede5efeb07b9adf9fbda7ea63f630d476a792

llama_cpp_python-0.2.85/.git/refs/tags/v0.2.85

@@ -0,0 +1 @@
+f7b9e6d42981bb75106699e0712a6378c39ef92c

llama_cpp_python-0.2.85/.git/shallow

@@ -0,0 +1 @@
+f7b9e6d42981bb75106699e0712a6378c39ef92c

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/CHANGELOG.md

@@ -7,6 +7,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.2.85]
+
+- feat: Update llama.cpp to ggerganov/llama.cpp@398ede5efeb07b9adf9fbda7ea63f630d476a792
+- fix: Missing LoRA adapter after API change by @shamitv in #1630
+- fix(docker): Update Dockerfile BLAS options by @olivierdebauche in #1632
+- fix(docker): Fix GGML_CUDA param by @olivierdebauche in #1633
+- fix(docker): Update Dockerfile build options from `LLAMA_` to `GGML_` by @olivierdebauche in #1634
+- feat: FreeBSD compatibility by @yurivict in #1635
+
 ## [0.2.84]
 
 - feat: Update llama.cpp to ggerganov/llama.cpp@4730faca618ff9cee0780580145e3cbe86f24876

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: llama_cpp_python
-Version: 0.2.84
+Version: 0.2.85
 Summary: Python bindings for the llama.cpp library
 Author-Email: Andrei Betlen <abetlen@gmail.com>
 License: MIT

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/docker/cuda_simple/Dockerfile

@@ -15,13 +15,13 @@ COPY . .
 
 # setting build related env vars
 ENV CUDA_DOCKER_ARCH=all
-ENV LLAMA_CUBLAS=1
+ENV GGML_CUDA=1
 
 # Install depencencies
 RUN python3 -m pip install --upgrade pip pytest cmake scikit-build setuptools fastapi uvicorn sse-starlette pydantic-settings starlette-context
 
 # Install llama-cpp-python (build with cuda)
-RUN CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
+RUN CMAKE_ARGS="-DGGML_CUDA=on" pip install llama-cpp-python
 
 # Run the server
 CMD python3 -m llama_cpp.server

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/docker/open_llama/Dockerfile

@@ -20,13 +20,13 @@ RUN python3 -m pip install --upgrade pip pytest cmake scikit-build setuptools fa
 
 # Perform the conditional installations based on the image
 RUN echo "Image: ${IMAGE}" && \
-    if [ "${IMAGE}" = "python:3-slim-bullseye" ] ; then \
+    if [ "${IMAGE}" = "python:3-slim-bookworm" ] ; then \
     echo "OpenBLAS install:" && \
     apt-get install -y --no-install-recommends libopenblas-dev && \
-    LLAMA_OPENBLAS=1 pip install llama-cpp-python --verbose; \
+    CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python --verbose; \
 else \
     echo "CuBLAS install:" && \
-    LLAMA_CUBLAS=1 pip install llama-cpp-python --verbose; \
+    CMAKE_ARGS="-DGGML_CUDA=on" pip install llama-cpp-python --verbose; \
 fi
 
 # Clean up apt cache

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/docker/openblas_simple/Dockerfile

@@ -12,7 +12,7 @@ RUN apt update && apt install -y libopenblas-dev ninja-build build-essential pkg
     
 RUN python -m pip install --upgrade pip pytest cmake scikit-build setuptools fastapi uvicorn sse-starlette pydantic-settings starlette-context
 
-RUN CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama_cpp_python --verbose
+RUN CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS" pip install llama_cpp_python --verbose
 
 # Run the server
 CMD python3 -m llama_cpp.server

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/llama_cpp/__init__.py

@@ -1,4 +1,4 @@
 from .llama_cpp import *
 from .llama import *
 
-__version__ = "0.2.84"
+__version__ = "0.2.85"

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/llama_cpp/llama.py

@@ -2083,11 +2083,14 @@ class Llama:
 
     def close(self) -> None:
         """Explicitly free the model from memory."""
-        self._stack.close()
+        if hasattr(self,'_stack'):
+            if self._stack is not None:
+                self._stack.close()
 
     def __del__(self) -> None:
-        if self._lora_adapter is not None:
-            llama_cpp.llama_lora_adapter_free(self._lora_adapter)
+        if hasattr(self,'_lora_adapter'):
+            if self._lora_adapter is not None:
+                llama_cpp.llama_lora_adapter_free(self._lora_adapter)
         self.close()
 
     @staticmethod

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/llama_cpp/llama_cpp.py

@@ -28,7 +28,7 @@ def _load_shared_library(lib_base_name: str):
     # for llamacpp) and "llama" (default name for this repo)
     _lib_paths: List[pathlib.Path] = []
     # Determine the file extension based on the platform
-    if sys.platform.startswith("linux"):
+    if sys.platform.startswith("linux") or sys.platform.startswith("freebsd"):
         _lib_paths += [
             _base_path / f"lib{lib_base_name}.so",
         ]

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/vendor/llama.cpp/.devops/nix/package.nix

@@ -126,16 +126,9 @@ let
     ++ optionals useMetalKit [ MetalKit ];
 
   cudaBuildInputs = with cudaPackages; [
-    cuda_cccl.dev # <nv/target>
-
-    # A temporary hack for reducing the closure size, remove once cudaPackages
-    # have stopped using lndir: https://github.com/NixOS/nixpkgs/issues/271792
-    cuda_cudart.dev
-    cuda_cudart.lib
-    cuda_cudart.static
-    libcublas.dev
-    libcublas.lib
-    libcublas.static
+    cuda_cudart
+    cuda_cccl # <nv/target>
+    libcublas
   ];
 
   rocmBuildInputs = with rocmPackages; [

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/vendor/llama.cpp/.github/workflows/build.yml

@@ -860,7 +860,8 @@ jobs:
           mkdir build
           cd build
           cmake .. -DGGML_NATIVE=OFF -DLLAMA_BUILD_SERVER=ON -DGGML_CUDA=ON -DBUILD_SHARED_LIBS=ON
-          cmake --build . --config Release -j $((${env:NUMBER_OF_PROCESSORS} - 1))
+          cmake --build . --config Release -j $((${env:NUMBER_OF_PROCESSORS} - 1)) -t ggml
+          cmake --build . --config Release -j ${env:NUMBER_OF_PROCESSORS}
 
       - name: Determine tag name
         id: tag

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/vendor/llama.cpp/CMakeLists.txt

@@ -139,7 +139,8 @@ set(LLAMA_BIN_INSTALL_DIR     ${CMAKE_INSTALL_BINDIR}     CACHE PATH "Location o
 # determining _precisely_ which defines are necessary for the llama-config
 # package.
 #
-get_directory_property(GGML_DIR_DEFINES DIRECTORY ggml/src COMPILE_DEFINITIONS)
+get_target_property(GGML_DIRECTORY ggml SOURCE_DIR)
+get_directory_property(GGML_DIR_DEFINES DIRECTORY ${GGML_DIRECTORY} COMPILE_DEFINITIONS)
 get_target_property(GGML_TARGET_DEFINES ggml COMPILE_DEFINITIONS)
 set(GGML_TRANSIENT_DEFINES ${GGML_TARGET_DEFINES} ${GGML_DIR_DEFINES})
 get_target_property(GGML_LINK_LIBRARIES ggml LINK_LIBRARIES)

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/vendor/llama.cpp/common/build-info.cpp

@@ -1,4 +1,4 @@
 int LLAMA_BUILD_NUMBER = 1;
-char const *LLAMA_COMMIT = "4730fac";
+char const *LLAMA_COMMIT = "398ede5";
 char const *LLAMA_COMPILER = "gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0";
 char const *LLAMA_BUILD_TARGET = "x86_64-linux-gnu";

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/vendor/llama.cpp/flake.lock

@@ -20,11 +20,11 @@
     },
     "nixpkgs": {
       "locked": {
-        "lastModified": 1721379653,
-        "narHash": "sha256-8MUgifkJ7lkZs3u99UDZMB4kbOxvMEXQZ31FO3SopZ0=",
+        "lastModified": 1722062969,
+        "narHash": "sha256-QOS0ykELUmPbrrUGmegAUlpmUFznDQeR4q7rFhl8eQg=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "1d9c2c9b3e71b9ee663d11c5d298727dace8d374",
+        "rev": "b73c2221a46c13557b1b3be9c2070cc42cf01eb3",
         "type": "github"
       },
       "original": {

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/vendor/llama.cpp/ggml/CMakeLists.txt

@@ -207,6 +207,7 @@ set(GGML_PUBLIC_HEADERS
     include/ggml-alloc.h
     include/ggml-backend.h
     include/ggml-blas.h
+    include/ggml-cann.h
     include/ggml-cuda.h
     include/ggml.h
     include/ggml-kompute.h

{llama_cpp_python-0.2.84 → llama_cpp_python-0.2.85}/vendor/llama.cpp/ggml/src/CMakeLists.txt

@@ -849,11 +849,6 @@ if (GGML_CANN)
                 ${CANN_INSTALL_DIR}/acllib/include
             )
 
-            # TODO: find libs
-            link_directories(
-                ${CANN_INSTALL_DIR}/lib64
-            )
-
             add_subdirectory(ggml-cann/kernels)
             list(APPEND CANN_LIBRARIES
                 ascendcl
@@ -872,6 +867,7 @@ if (GGML_CANN)
 
             set(GGML_EXTRA_LIBS     ${GGML_EXTRA_LIBS}     ${CANN_LIBRARIES} )
             set(GGML_EXTRA_INCLUDES ${GGML_EXTRA_INCLUDES} ${CANN_INCLUDE_DIRS})
+            set(GGML_EXTRA_LIBDIRS  ${GGML_EXTRA_LIBDIRS}  ${CANN_INSTALL_DIR}/lib64)
             list(APPEND GGML_CDEF_PUBLIC GGML_USE_CANN)
         endif()
     else()