Sync abetlen/llama-cpp-python 0.3.12 code

Author: JamePeng

.github/workflows/build-and-release.yaml

@@ -74,6 +74,7 @@ jobs:
           CIBW_SKIP: "*musllinux* pp*"
           CIBW_REPAIR_WHEEL_COMMAND: ""
           CIBW_ARCHS: "aarch64"
+          CIBW_ENVIRONMENT: CMAKE_ARGS="-DCMAKE_OSX_ARCHITECTURES=arm64 -DCMAKE_APPLE_SILICON_PROCESSOR=arm64 -DCMAKE_CROSSCOMPILING=ON"
           CIBW_BUILD: "cp38-* cp39-* cp310-* cp311-* cp312-*"
         with:
           output-dir: wheelhouse

.github/workflows/build-docker.yaml

@@ -9,7 +9,7 @@ permissions:
 jobs:
   docker:
     name: Build and push Docker image
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     steps:
       - name: Checkout
         uses: actions/checkout@v4

CHANGELOG.md

@@ -7,6 +7,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.3.12]
+
+- feat: Update llama.cpp to [ggml-org/llama.cpp@b8eeb8741d4483daf576498cf90537b4de71633c](https://github.com/ggml-org/llama.cpp/commit/b8eeb8741d4483daf576498cf90537b4de71633c)
+
 ## [0.3.11]
 
 - fix: Update reference to `llama_kv_cache_clear` in Llama.embed. Closes #2037 by @abetlen in 9e5a4eaa84156084ed7bbb91e6efcc91dc6217bc

llama_cpp/__init__.py

@@ -1,4 +1,4 @@
 from .llama_cpp import *
 from .llama import *
 
-__version__ = "0.3.11"
+__version__ = "0.3.12"