llama-cpp-python 0.3.10__tar.gz → 0.3.11__tar.gz

llama_cpp_python-0.3.11/.git/FETCH_HEAD

@@ -0,0 +1 @@
+1580839fe8fa01324a0d33bc52d63279df368eec		'1580839fe8fa01324a0d33bc52d63279df368eec' of https://github.com/abetlen/llama-cpp-python

llama_cpp_python-0.3.11/.git/HEAD

@@ -0,0 +1 @@
+1580839fe8fa01324a0d33bc52d63279df368eec

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/.git/config

@@ -9,7 +9,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX0QxZ0E3R25OY3BzVU00WWlsTFppOUpHcVltNFAwdzFsVVRsUg==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX3RhbUVPYVRXU2g3TDU2RGdXeHNzN2FWQWFIQzJzTDE0T0dHTQ==
 [submodule "vendor/llama.cpp"]
 	active = true
 	url = https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.3.11/.git/logs/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 1580839fe8fa01324a0d33bc52d63279df368eec runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696788 +0000	checkout: moving from master to refs/tags/v0.3.11

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/.git/modules/vendor/llama.cpp/config

@@ -16,7 +16,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX0QxZ0E3R25OY3BzVU00WWlsTFppOUpHcVltNFAwdzFsVVRsUg==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX3RhbUVPYVRXU2g3TDU2RGdXeHNzN2FWQWFIQzJzTDE0T0dHTQ==
 [url "https://github.com/"]
 	insteadOf = git@github.com:
 	insteadOf = org-6826477@github.com:

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/logs/HEAD

@@ -0,0 +1,2 @@
+0000000000000000000000000000000000000000 668168814601d2aef6161ce49ab186bc74177ae7 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696790 +0000	clone: from https://github.com/ggerganov/llama.cpp.git
+668168814601d2aef6161ce49ab186bc74177ae7 8846aace4934ad29651ea61b8c7e3f6b0556e3d2 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696792 +0000	checkout: moving from master to 8846aace4934ad29651ea61b8c7e3f6b0556e3d2

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/logs/refs/heads/master

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 668168814601d2aef6161ce49ab186bc74177ae7 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696790 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/logs/refs/remotes/origin/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 668168814601d2aef6161ce49ab186bc74177ae7 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696790 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/.git/modules/vendor/llama.cpp/modules/kompute/config

@@ -13,7 +13,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX0QxZ0E3R25OY3BzVU00WWlsTFppOUpHcVltNFAwdzFsVVRsUg==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzX3RhbUVPYVRXU2g3TDU2RGdXeHNzN2FWQWFIQzJzTDE0T0dHTQ==
 [url "https://github.com/"]
 	insteadOf = git@github.com:
 	insteadOf = org-6826477@github.com:

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/modules/kompute/logs/HEAD

@@ -0,0 +1,2 @@
+0000000000000000000000000000000000000000 7c20efa30bb53d08bf04f84e510275766ebe9923 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696793 +0000	clone: from https://github.com/nomic-ai/kompute.git
+7c20efa30bb53d08bf04f84e510275766ebe9923 4565194ed7c32d1d2efa32ceab4d3c6cae006306 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696794 +0000	checkout: moving from master to 4565194ed7c32d1d2efa32ceab4d3c6cae006306

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/heads/master

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 7c20efa30bb53d08bf04f84e510275766ebe9923 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696793 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/remotes/origin/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 7c20efa30bb53d08bf04f84e510275766ebe9923 runner <runner@pkrvmbietmlfzoi.isbtygbhkgwune5ckp3zdudqla.dx.internal.cloudapp.net> 1751696793 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/packed-refs

@@ -0,0 +1,2 @@
+# pack-refs with: peeled fully-peeled sorted 
+668168814601d2aef6161ce49ab186bc74177ae7 refs/remotes/origin/master

llama_cpp_python-0.3.11/.git/modules/vendor/llama.cpp/refs/heads/master

@@ -0,0 +1 @@
+668168814601d2aef6161ce49ab186bc74177ae7

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/.git/modules/vendor/llama.cpp/shallow

@@ -1,2 +1,2 @@
+668168814601d2aef6161ce49ab186bc74177ae7
 8846aace4934ad29651ea61b8c7e3f6b0556e3d2
-c8c4495b8d3a8799e2d46778f993965b0ac1ae43

llama_cpp_python-0.3.11/.git/refs/tags/v0.3.11

@@ -0,0 +1 @@
+1580839fe8fa01324a0d33bc52d63279df368eec

llama_cpp_python-0.3.11/.git/shallow

@@ -0,0 +1 @@
+1580839fe8fa01324a0d33bc52d63279df368eec

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/.github/workflows/build-wheels-cuda.yaml

@@ -20,7 +20,7 @@ jobs:
         id: set-matrix
         run: |
           $matrix = @{
-              'os' = @('ubuntu-latest', 'windows-2019')
+              'os' = @('ubuntu-22.04', 'windows-2022')
               'pyver' = @("3.9", "3.10", "3.11", "3.12")
               'cuda' = @("12.1.1", "12.2.2", "12.3.2", "12.4.1") #, "12.5.1", "12.6.1")
               'releasetag' = @("basic")

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/.github/workflows/build-wheels-metal.yaml

@@ -11,7 +11,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [macos-13, macos-14, macos-15]
+        os: [macos-14, macos-15]
 
     steps:
       - uses: actions/checkout@v4
@@ -23,32 +23,21 @@ jobs:
         with:
           python-version: "3.12"
           cache: 'pip'
-          
+
       - name: Install dependencies (Linux/MacOS)
-        if: runner.os != 'Windows'
         run: |
           python -m pip install --upgrade pip
           python -m pip install uv
           RUST_LOG=trace python -m uv pip install -e .[all] --verbose
         shell: bash
 
-      - name: Install dependencies (Windows)
-        if: runner.os == 'Windows'
-        env:
-          RUST_LOG: trace        
-        run: |
-          python -m pip install --upgrade pip
-          python -m pip install uv
-          python -m uv pip install -e .[all] --verbose
-        shell: cmd
-
       - name: Build wheels
         uses: pypa/cibuildwheel@v2.22.0
         env:
           # disable repair
           CIBW_REPAIR_WHEEL_COMMAND: ""
           CIBW_ARCHS: "arm64"
-          CIBW_ENVIRONMENT: CMAKE_ARGS="-DCMAKE_OSX_ARCHITECTURES=arm64 -DCMAKE_APPLE_SILICON_PROCESSOR=arm64 -DGGML_METAL=on"
+          CIBW_ENVIRONMENT: CMAKE_ARGS="-DCMAKE_OSX_ARCHITECTURES=arm64 -DCMAKE_APPLE_SILICON_PROCESSOR=arm64 -DGGML_METAL=on -DCMAKE_CROSSCOMPILING=ON"
           CIBW_BUILD: "cp39-* cp310-* cp311-* cp312-*"
         with:
           package-dir: .
@@ -69,7 +58,7 @@ jobs:
         with:
           merge-multiple: true
           path: dist2
-          
+
       - uses: softprops/action-gh-release@v2
         with:
           files: dist2/*

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/CHANGELOG.md

@@ -7,6 +7,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.3.11]
+
+- fix: Update reference to `llama_kv_cache_clear` in Llama.embed. Closes #2037 by @abetlen in 9e5a4eaa84156084ed7bbb91e6efcc91dc6217bc
+
 ## [0.3.10]
 
 - feat: Update llama.cpp to ggerganov/llama.cpp@8846aace4934ad29651ea61b8c7e3f6b0556e3d2

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: llama_cpp_python
-Version: 0.3.10
+Version: 0.3.11
 Summary: Python bindings for the llama.cpp library
 Author-Email: Andrei Betlen <abetlen@gmail.com>
 License: MIT

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/docker/simple/Dockerfile

@@ -9,6 +9,7 @@ ARG IMAGE
 
 # Update and upgrade the existing packages 
 RUN apt-get update && apt-get upgrade -y && apt-get install -y --no-install-recommends \
+    git \
     python3 \
     python3-pip \
     ninja-build \

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/llama_cpp/__init__.py

@@ -1,4 +1,4 @@
 from .llama_cpp import *
 from .llama import *
 
-__version__ = "0.3.10"
+__version__ = "0.3.11"

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/llama_cpp/llama.py

@@ -1041,7 +1041,7 @@ class Llama:
         data: Union[List[List[float]], List[List[List[float]]]] = []
 
         def decode_batch(seq_sizes: List[int]):
-            llama_cpp.llama_kv_cache_clear(self._ctx.ctx)
+            llama_cpp.llama_kv_self_clear(self._ctx.ctx)
             self._ctx.decode(self._batch)
             self._batch.reset()
 
@@ -1112,7 +1112,7 @@ class Llama:
 
         output = data[0] if isinstance(input, str) else data
 
-        llama_cpp.llama_kv_cache_clear(self._ctx.ctx)
+        llama_cpp.llama_kv_self_clear(self._ctx.ctx)
         self.reset()
 
         if return_count:

{llama_cpp_python-0.3.10 → llama_cpp_python-0.3.11}/tests/test_llama.py

@@ -216,3 +216,19 @@ root ::= "1" | "2" | "3" | "4" | "5" | "6" | "7" | "8" | "9" | "10"
 
     assert number_1 != number_2
     assert number_1 == number_3
+
+
+def test_real_llama_embeddings(llama_cpp_model_path):
+    model = llama_cpp.Llama(
+        llama_cpp_model_path,
+        n_ctx=32,
+        n_batch=32,
+        n_ubatch=32,
+        n_threads=multiprocessing.cpu_count(),
+        n_threads_batch=multiprocessing.cpu_count(),
+        logits_all=False,
+        flash_attn=True,
+        embedding=True
+    )
+    # Smoke test for now
+    model.embed("Hello World")

llama_cpp_python-0.3.10/.git/FETCH_HEAD

@@ -1 +0,0 @@
-9770b8405fb05ed2599dcd3941e4db54d861c12e		'9770b8405fb05ed2599dcd3941e4db54d861c12e' of https://github.com/abetlen/llama-cpp-python

llama_cpp_python-0.3.10/.git/HEAD

@@ -1 +0,0 @@
-9770b8405fb05ed2599dcd3941e4db54d861c12e

llama_cpp_python-0.3.10/.git/logs/HEAD

@@ -1 +0,0 @@
-0000000000000000000000000000000000000000 9770b8405fb05ed2599dcd3941e4db54d861c12e runner <runner@pkrvmbietmlfzoi.(none)> 1751557077 +0000	checkout: moving from master to refs/tags/v0.3.10

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/logs/HEAD

@@ -1,2 +0,0 @@
-0000000000000000000000000000000000000000 c8c4495b8d3a8799e2d46778f993965b0ac1ae43 runner <runner@pkrvmbietmlfzoi.(none)> 1751557079 +0000	clone: from https://github.com/ggerganov/llama.cpp.git
-c8c4495b8d3a8799e2d46778f993965b0ac1ae43 8846aace4934ad29651ea61b8c7e3f6b0556e3d2 runner <runner@pkrvmbietmlfzoi.(none)> 1751557080 +0000	checkout: moving from master to 8846aace4934ad29651ea61b8c7e3f6b0556e3d2

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/logs/refs/heads/master

@@ -1 +0,0 @@
-0000000000000000000000000000000000000000 c8c4495b8d3a8799e2d46778f993965b0ac1ae43 runner <runner@pkrvmbietmlfzoi.(none)> 1751557079 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/logs/refs/remotes/origin/HEAD

@@ -1 +0,0 @@
-0000000000000000000000000000000000000000 c8c4495b8d3a8799e2d46778f993965b0ac1ae43 runner <runner@pkrvmbietmlfzoi.(none)> 1751557079 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/modules/kompute/logs/HEAD

@@ -1,2 +0,0 @@
-0000000000000000000000000000000000000000 7c20efa30bb53d08bf04f84e510275766ebe9923 runner <runner@pkrvmbietmlfzoi.(none)> 1751557081 +0000	clone: from https://github.com/nomic-ai/kompute.git
-7c20efa30bb53d08bf04f84e510275766ebe9923 4565194ed7c32d1d2efa32ceab4d3c6cae006306 runner <runner@pkrvmbietmlfzoi.(none)> 1751557082 +0000	checkout: moving from master to 4565194ed7c32d1d2efa32ceab4d3c6cae006306

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/heads/master

@@ -1 +0,0 @@
-0000000000000000000000000000000000000000 7c20efa30bb53d08bf04f84e510275766ebe9923 runner <runner@pkrvmbietmlfzoi.(none)> 1751557081 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/remotes/origin/HEAD

@@ -1 +0,0 @@
-0000000000000000000000000000000000000000 7c20efa30bb53d08bf04f84e510275766ebe9923 runner <runner@pkrvmbietmlfzoi.(none)> 1751557081 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/packed-refs

@@ -1,2 +0,0 @@
-# pack-refs with: peeled fully-peeled sorted 
-c8c4495b8d3a8799e2d46778f993965b0ac1ae43 refs/remotes/origin/master

llama_cpp_python-0.3.10/.git/modules/vendor/llama.cpp/refs/heads/master

@@ -1 +0,0 @@
-c8c4495b8d3a8799e2d46778f993965b0ac1ae43

llama_cpp_python-0.3.10/.git/refs/tags/v0.3.10

@@ -1 +0,0 @@
-9770b8405fb05ed2599dcd3941e4db54d861c12e

llama_cpp_python-0.3.10/.git/shallow

@@ -1 +0,0 @@
-9770b8405fb05ed2599dcd3941e4db54d861c12e