llama-cpp-python 0.2.59__tar.gz → 0.2.60__tar.gz

llama_cpp_python-0.2.60/.git/FETCH_HEAD

@@ -0,0 +1 @@
+08b16afe11e7b42adec2fed0a781123383476045		'08b16afe11e7b42adec2fed0a781123383476045' of https://github.com/abetlen/llama-cpp-python

llama_cpp_python-0.2.60/.git/HEAD

@@ -0,0 +1 @@
+08b16afe11e7b42adec2fed0a781123383476045

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/.git/config

@@ -9,7 +9,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzXzRPWHJKQ0tPeTR3Y25uYWVBMmF3VE9IMFhtZzNOVTJCM3hYeQ==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzXzVMZFpGUjhRNkxlRTRIUkZSUlg1cjM0QU1zY2hzcTBjTDg3Nw==
 [submodule "vendor/llama.cpp"]
 	active = true
 	url = https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.2.60/.git/logs/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 08b16afe11e7b42adec2fed0a781123383476045 runner <runner@fv-az768-866.(none)> 1712383168 +0000	checkout: moving from master to refs/tags/v0.2.60

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/HEAD

@@ -0,0 +1 @@
+75cd4c77292034ecec587ecb401366f57338f7c0

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/.git/modules/vendor/llama.cpp/config

@@ -16,7 +16,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzXzRPWHJKQ0tPeTR3Y25uYWVBMmF3VE9IMFhtZzNOVTJCM3hYeQ==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzXzVMZFpGUjhRNkxlRTRIUkZSUlg1cjM0QU1zY2hzcTBjTDg3Nw==
 [url "https://github.com/"]
 	insteadOf = git@github.com:
 	insteadOf = org-6826477@github.com:

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/logs/HEAD

@@ -0,0 +1,2 @@
+0000000000000000000000000000000000000000 75cd4c77292034ecec587ecb401366f57338f7c0 runner <runner@fv-az768-866.(none)> 1712383169 +0000	clone: from https://github.com/ggerganov/llama.cpp.git
+75cd4c77292034ecec587ecb401366f57338f7c0 75cd4c77292034ecec587ecb401366f57338f7c0 runner <runner@fv-az768-866.(none)> 1712383169 +0000	checkout: moving from master to 75cd4c77292034ecec587ecb401366f57338f7c0

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/logs/refs/heads/master

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 75cd4c77292034ecec587ecb401366f57338f7c0 runner <runner@fv-az768-866.(none)> 1712383169 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/logs/refs/remotes/origin/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 75cd4c77292034ecec587ecb401366f57338f7c0 runner <runner@fv-az768-866.(none)> 1712383169 +0000	clone: from https://github.com/ggerganov/llama.cpp.git

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/.git/modules/vendor/llama.cpp/modules/kompute/config

@@ -13,7 +13,7 @@
 [gc]
 	auto = 0
 [http "https://github.com/"]
-	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzXzRPWHJKQ0tPeTR3Y25uYWVBMmF3VE9IMFhtZzNOVTJCM3hYeQ==
+	extraheader = AUTHORIZATION: basic eC1hY2Nlc3MtdG9rZW46Z2hzXzVMZFpGUjhRNkxlRTRIUkZSUlg1cjM0QU1zY2hzcTBjTDg3Nw==
 [url "https://github.com/"]
 	insteadOf = git@github.com:
 	insteadOf = org-6826477@github.com:

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/modules/kompute/logs/HEAD

@@ -0,0 +1,2 @@
+0000000000000000000000000000000000000000 d1e3b0953cf66acc94b2e29693e221427b2c1f3f runner <runner@fv-az768-866.(none)> 1712383170 +0000	clone: from https://github.com/nomic-ai/kompute.git
+d1e3b0953cf66acc94b2e29693e221427b2c1f3f 4565194ed7c32d1d2efa32ceab4d3c6cae006306 runner <runner@fv-az768-866.(none)> 1712383171 +0000	checkout: moving from master to 4565194ed7c32d1d2efa32ceab4d3c6cae006306

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/heads/master

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 d1e3b0953cf66acc94b2e29693e221427b2c1f3f runner <runner@fv-az768-866.(none)> 1712383170 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/modules/kompute/logs/refs/remotes/origin/HEAD

@@ -0,0 +1 @@
+0000000000000000000000000000000000000000 d1e3b0953cf66acc94b2e29693e221427b2c1f3f runner <runner@fv-az768-866.(none)> 1712383170 +0000	clone: from https://github.com/nomic-ai/kompute.git

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/packed-refs

@@ -0,0 +1,2 @@
+# pack-refs with: peeled fully-peeled sorted 
+75cd4c77292034ecec587ecb401366f57338f7c0 refs/remotes/origin/master

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/refs/heads/master

@@ -0,0 +1 @@
+75cd4c77292034ecec587ecb401366f57338f7c0

llama_cpp_python-0.2.60/.git/modules/vendor/llama.cpp/shallow

@@ -0,0 +1 @@
+75cd4c77292034ecec587ecb401366f57338f7c0

llama_cpp_python-0.2.60/.git/refs/tags/v0.2.60

@@ -0,0 +1 @@
+08b16afe11e7b42adec2fed0a781123383476045

llama_cpp_python-0.2.60/.git/shallow

@@ -0,0 +1 @@
+08b16afe11e7b42adec2fed0a781123383476045

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/CHANGELOG.md

@@ -7,6 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.2.60]
+
+- feat: Update llama.cpp to ggerganov/llama.cpp@75cd4c77292034ecec587ecb401366f57338f7c0
+- fix: Always embed metal library by @abetlen in b3bfea6dbfb6ed9ce18f9a2723e0a9e4bd1da7ad
+- fix: missing logprobs in response, incorrect response type for functionary by @abetlen in 1ae3abbcc3af7f4a25a3ffc40b246f18039565e8
+- fix(docs): incorrect tool_choice example by @CISC in #1330
+
 ## [0.2.59]
 
 - feat: Update llama.cpp to ggerganov/llama.cpp@ba0c7c70ab5b15f1f2be7fb0dfbe0366dda30d6c

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/CMakeLists.txt

@@ -18,7 +18,7 @@ if (LLAMA_BUILD)
         set(LLAMA_F16C "Off" CACHE BOOL "llama: enable F16C" FORCE)
     endif()
 
-    if (APPLE AND CMAKE_SYSTEM_PROCESSOR MATCHES "arm64")
+    if (APPLE)
         set(LLAMA_METAL_EMBED_LIBRARY "On" CACHE BOOL "llama: embed metal library" FORCE)
     endif()
 

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: llama_cpp_python
-Version: 0.2.59
+Version: 0.2.60
 Summary: Python bindings for the llama.cpp library
 Author-Email: Andrei Betlen <abetlen@gmail.com>
 License: MIT
@@ -87,6 +87,15 @@ This will also build `llama.cpp` from source and install it alongside this pytho
 
 If this fails, add `--verbose` to the `pip install` see the full cmake build log.
 
+**Pre-built Wheel (New)**
+
+It is also possible to install a pre-built wheel with basic CPU support.
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
+```
+
 ### Installation Configuration
 
 `llama.cpp` supports a number of hardware acceleration backends to speed up inference as well as backend specific options. See the [llama.cpp README](https://github.com/ggerganov/llama.cpp#build) for a full list.
@@ -143,12 +152,36 @@ CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-
 </details>
 
 <details>
-<summary>cuBLAS (CUDA)</summary>
+<summary>CUDA</summary>
+
+To install with CUDA support, set the `LLAMA_CUDA=on` environment variable before installing:
+
+```bash
+CMAKE_ARGS="-DLLAMA_CUDA=on" pip install llama-cpp-python
+```
+
+**Pre-built Wheel (New)**
+
+It is also possible to install a pre-built wheel with CUDA support. As long as your system meets some requirements:
+
+- CUDA Version is 12.1, 12.2 or 12.3
+- Python Version is 3.10, 3.11 or 3.12
 
-To install with cuBLAS, set the `LLAMA_CUBLAS=on` environment variable before installing:
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/<cuda-version>
+```
+
+Where `<cuda-version>` is one of the following:
+- `cu121`: CUDA 12.1
+- `cu122`: CUDA 12.2
+- `cu123`: CUDA 12.3
+
+For example, to install the CUDA 12.1 wheel:
 
 ```bash
-CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu121
 ```
 
 </details>
@@ -162,6 +195,18 @@ To install with Metal (MPS), set the `LLAMA_METAL=on` environment variable befor
 CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
 ```
 
+**Pre-built Wheel (New)**
+
+It is also possible to install a pre-built wheel with Metal support. As long as your system meets some requirements:
+
+- MacOS Version is 11.0 or later
+- Python Version is 3.10, 3.11 or 3.12
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/metal
+```
+
 </details>
 <details>
 
@@ -456,12 +501,12 @@ The high-level API supports OpenAI compatible function and tool calling. This is
           }
         }
       }],
-      tool_choice=[{
+      tool_choice={
         "type": "function",
         "function": {
           "name": "UserDetail"
         }
-      }]
+      }
 )
 ```
 
@@ -612,7 +657,7 @@ python3 -m llama_cpp.server --model models/7B/llama-model.gguf
 Similar to Hardware Acceleration section above, you can also install with GPU (cuBLAS) support like this:
 
 ```bash
-CMAKE_ARGS="-DLLAMA_CUBLAS=on" FORCE_CMAKE=1 pip install 'llama-cpp-python[server]'
+CMAKE_ARGS="-DLLAMA_CUDA=on" FORCE_CMAKE=1 pip install 'llama-cpp-python[server]'
 python3 -m llama_cpp.server --model models/7B/llama-model.gguf --n_gpu_layers 35
 ```
 

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/README.md

@@ -44,6 +44,15 @@ This will also build `llama.cpp` from source and install it alongside this pytho
 
 If this fails, add `--verbose` to the `pip install` see the full cmake build log.
 
+**Pre-built Wheel (New)**
+
+It is also possible to install a pre-built wheel with basic CPU support.
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
+```
+
 ### Installation Configuration
 
 `llama.cpp` supports a number of hardware acceleration backends to speed up inference as well as backend specific options. See the [llama.cpp README](https://github.com/ggerganov/llama.cpp#build) for a full list.
@@ -100,12 +109,36 @@ CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-
 </details>
 
 <details>
-<summary>cuBLAS (CUDA)</summary>
+<summary>CUDA</summary>
+
+To install with CUDA support, set the `LLAMA_CUDA=on` environment variable before installing:
+
+```bash
+CMAKE_ARGS="-DLLAMA_CUDA=on" pip install llama-cpp-python
+```
+
+**Pre-built Wheel (New)**
+
+It is also possible to install a pre-built wheel with CUDA support. As long as your system meets some requirements:
+
+- CUDA Version is 12.1, 12.2 or 12.3
+- Python Version is 3.10, 3.11 or 3.12
 
-To install with cuBLAS, set the `LLAMA_CUBLAS=on` environment variable before installing:
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/<cuda-version>
+```
+
+Where `<cuda-version>` is one of the following:
+- `cu121`: CUDA 12.1
+- `cu122`: CUDA 12.2
+- `cu123`: CUDA 12.3
+
+For example, to install the CUDA 12.1 wheel:
 
 ```bash
-CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu121
 ```
 
 </details>
@@ -119,6 +152,18 @@ To install with Metal (MPS), set the `LLAMA_METAL=on` environment variable befor
 CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
 ```
 
+**Pre-built Wheel (New)**
+
+It is also possible to install a pre-built wheel with Metal support. As long as your system meets some requirements:
+
+- MacOS Version is 11.0 or later
+- Python Version is 3.10, 3.11 or 3.12
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/metal
+```
+
 </details>
 <details>
 
@@ -413,12 +458,12 @@ The high-level API supports OpenAI compatible function and tool calling. This is
           }
         }
       }],
-      tool_choice=[{
+      tool_choice={
         "type": "function",
         "function": {
           "name": "UserDetail"
         }
-      }]
+      }
 )
 ```
 
@@ -569,7 +614,7 @@ python3 -m llama_cpp.server --model models/7B/llama-model.gguf
 Similar to Hardware Acceleration section above, you can also install with GPU (cuBLAS) support like this:
 
 ```bash
-CMAKE_ARGS="-DLLAMA_CUBLAS=on" FORCE_CMAKE=1 pip install 'llama-cpp-python[server]'
+CMAKE_ARGS="-DLLAMA_CUDA=on" FORCE_CMAKE=1 pip install 'llama-cpp-python[server]'
 python3 -m llama_cpp.server --model models/7B/llama-model.gguf --n_gpu_layers 35
 ```
 

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/llama_cpp/__init__.py

@@ -1,4 +1,4 @@
 from .llama_cpp import *
 from .llama import *
 
-__version__ = "0.2.59"
+__version__ = "0.2.60"

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/llama_cpp/llama_chat_format.py

@@ -6,7 +6,7 @@ import ctypes
 import dataclasses
 import random
 import string
-from typing import Any, Dict, Iterator, List, Literal, Optional, Tuple, Union, Protocol
+from typing import Any, Dict, Iterator, List, Literal, Optional, Tuple, Union, Protocol, cast
 
 import jinja2
 
@@ -338,6 +338,7 @@ def _convert_completion_to_chat_function(
                             }
                         ],
                     },
+                    "logprobs": None,
                     "finish_reason": "tool_calls",
                 }
             ],
@@ -1191,7 +1192,6 @@ def format_mistral_instruct(
         elif (
             message["role"] == "assistant"
             and message["content"] is not None
-            and isinstance(message["content"], str)
         ):
             prompt += " [/INST]" + message["content"] + eos
     prompt += " [/INST]"
@@ -1263,7 +1263,7 @@ def format_gemma(
     **kwargs: Any,
 ) -> ChatFormatterResponse:
     system_message = _get_system_message(messages)
-    if system_message is not None and system_message != "":
+    if system_message != "":
         logger.debug(
             "`role='system'` messages are not allowed on Google's Gemma models."
         )
@@ -1628,6 +1628,7 @@ def functionary_chat_handler(
                         }
                     ],
                 },
+                "logprobs": None,
                 "finish_reason": "tool_calls",
             }
         ],
@@ -1909,14 +1910,14 @@ def functionary_v1_v2_chat_handler(
         return grammar
 
     def create_completion(stop):
-        completion: llama_types.Completion = llama.create_completion(
+        completion = cast(llama_types.Completion, llama.create_completion(
             prompt=prompt,
             temperature=temperature,
             top_p=top_p,
             top_k=top_k,
             min_p=min_p,
             typical_p=typical_p,
-            stream=stream,
+            stream=False,
             stop=stop,
             max_tokens=max_tokens,
             presence_penalty=presence_penalty,
@@ -1929,7 +1930,7 @@ def functionary_v1_v2_chat_handler(
             model=model,
             logits_processor=logits_processor,
             grammar=grammar,
-        )
+        ))
 
         return completion
 
@@ -2050,7 +2051,7 @@ def functionary_v1_v2_chat_handler(
     assert "usage" in completion
     assert len(function_calls) == len(function_bodies)
 
-    tool_calls = []
+    tool_calls: List[llama_types.ChatCompletionMessageToolCall] = []
     for function_call, function_body in zip(function_calls, function_bodies):
         tool_calls.append(
             {
@@ -2070,6 +2071,12 @@ def functionary_v1_v2_chat_handler(
         )
 
     # TODO: support stream mode
+    function_call_dict: Union[Dict[str, str], Dict[Literal["function_call"], llama_types.ChatCompletionRequestAssistantMessageFunctionCall]] = { 
+        "function_call": {
+            "name": tool_calls[0]["function"]["name"],
+            "arguments": tool_calls[0]["function"]["arguments"],
+        }
+    } if len(tool_calls) == 1 else {}
     return llama_types.CreateChatCompletionResponse(
         id="chat" + completion["id"],
         object="chat.completion",
@@ -2078,14 +2085,12 @@ def functionary_v1_v2_chat_handler(
         choices=[
             {
                 "index": 0,
+                "logprobs": None,
                 "message": {
                     "role": "assistant",
                     "content": None if content == "" else content,
-                    "function_call": {
-                        "name": tool_calls[0]["function"]["name"],
-                        "arguments": tool_calls[0]["function"]["arguments"],
-                    } if len(tool_calls) > 0 else None,
-                    "tool_calls": tool_calls if len(tool_calls) > 0 else None,
+                    "tool_calls": tool_calls,
+                    **function_call_dict,
                 },
                 "finish_reason": "tool_calls" if len(tool_calls) > 0 else "stop",
             }
@@ -2565,8 +2570,8 @@ def chatml_function_calling(
     tool_name = text[len("functions.") :]
     tool = next((tool for tool in tools if tool["function"]["name"] == tool_name), None)
     if not stream:
-        completions = []
-        completions_tool_name = []
+        completions: List[llama_types.CreateCompletionResponse] = []
+        completions_tool_name: List[str] = []
         while tool is not None:
             prompt += f"functions.{tool_name}:\n"
             try:
@@ -2603,6 +2608,7 @@ def chatml_function_calling(
                 logits_processor=logits_processor,
                 grammar=grammar,
             )
+            completion_or_chunks = cast(llama_types.CreateCompletionResponse, completion_or_chunks)
             completions.append(completion_or_chunks)
             completions_tool_name.append(tool_name)
             prompt += completion_or_chunks["choices"][0]["text"]
@@ -2631,6 +2637,7 @@ def chatml_function_calling(
                     follow_up_gbnf_tool_grammar, verbose=llama.verbose
                 ),
             )
+            response = cast(llama_types.CreateCompletionResponse, response)
 
             tool_name = response["choices"][0]["text"][len("functions.") :]
             tool = next(
@@ -2638,7 +2645,7 @@ def chatml_function_calling(
             )
 
         # Merge completions
-        function_call = { 
+        function_call_dict: Union[Dict[str, str], Dict[Literal["function_call"], llama_types.ChatCompletionRequestAssistantMessageFunctionCall]] = { 
             "function_call": {
                 "name": tool_name,
                 "arguments": completions[0]["choices"][0]["text"],
@@ -2653,6 +2660,7 @@ def chatml_function_calling(
                 {
                     "finish_reason": "tool_calls",
                     "index": 0,
+                    "logprobs": None,
                     "message": {
                         "role": "assistant",
                         "content": None,
@@ -2673,20 +2681,22 @@ def chatml_function_calling(
                                 zip(completions_tool_name, completions)
                             )
                         ],
-                        **function_call
+                        **function_call_dict
                     },
                 }
             ],
             "usage": {
                 "completion_tokens": sum(
-                    completion["usage"]["completion_tokens"]
+                    completion["usage"]["completion_tokens"] if "usage" in completion else 0
                     for completion in completions
                 ),
                 "prompt_tokens": sum(
-                    completion["usage"]["prompt_tokens"] for completion in completions
+                    completion["usage"]["prompt_tokens"] if "usage" in completion else 0
+                    for completion in completions
                 ),
                 "total_tokens": sum(
-                    completion["usage"]["total_tokens"] for completion in completions
+                    completion["usage"]["total_tokens"] if "usage" in completion else 0
+                    for completion in completions
                 ),
             },
         }

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/vendor/llama.cpp/.devops/full-cuda.Dockerfile

@@ -12,7 +12,7 @@ FROM ${BASE_CUDA_DEV_CONTAINER} as build
 ARG CUDA_DOCKER_ARCH=all
 
 RUN apt-get update && \
-    apt-get install -y build-essential python3 python3-pip git
+    apt-get install -y build-essential python3 python3-pip git libcurl4-openssl-dev
 
 COPY requirements.txt   requirements.txt
 COPY requirements       requirements
@@ -28,6 +28,8 @@ COPY . .
 ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH}
 # Enable CUDA
 ENV LLAMA_CUDA=1
+# Enable cURL
+ENV LLAMA_CURL=1
 
 RUN make
 

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/vendor/llama.cpp/.devops/full-rocm.Dockerfile

@@ -40,6 +40,11 @@ ENV LLAMA_HIPBLAS=1
 ENV CC=/opt/rocm/llvm/bin/clang
 ENV CXX=/opt/rocm/llvm/bin/clang++
 
+# Enable cURL
+ENV LLAMA_CURL=1
+RUN apt-get update && \
+    apt-get install -y libcurl4-openssl-dev
+
 RUN make
 
 ENTRYPOINT ["/app/.devops/tools.sh"]

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/vendor/llama.cpp/.devops/full.Dockerfile

@@ -3,7 +3,7 @@ ARG UBUNTU_VERSION=22.04
 FROM ubuntu:$UBUNTU_VERSION as build
 
 RUN apt-get update && \
-    apt-get install -y build-essential python3 python3-pip git
+    apt-get install -y build-essential python3 python3-pip git libcurl4-openssl-dev
 
 COPY requirements.txt   requirements.txt
 COPY requirements       requirements
@@ -15,6 +15,9 @@ WORKDIR /app
 
 COPY . .
 
+ENV LLAMA_CURL=1
+
+
 RUN make
 
 ENV LC_ALL=C.utf8

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/vendor/llama.cpp/.devops/server-cuda.Dockerfile

@@ -12,7 +12,7 @@ FROM ${BASE_CUDA_DEV_CONTAINER} as build
 ARG CUDA_DOCKER_ARCH=all
 
 RUN apt-get update && \
-    apt-get install -y build-essential git
+    apt-get install -y build-essential git libcurl4-openssl-dev
 
 WORKDIR /app
 
@@ -22,11 +22,16 @@ COPY . .
 ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH}
 # Enable CUDA
 ENV LLAMA_CUDA=1
+# Enable cURL
+ENV LLAMA_CURL=1
 
 RUN make
 
 FROM ${BASE_CUDA_RUN_CONTAINER} as runtime
 
+RUN apt-get update && \
+    apt-get install -y libcurl4-openssl-dev
+
 COPY --from=build /app/server /server
 
 ENTRYPOINT [ "/server" ]

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/vendor/llama.cpp/.devops/server-intel.Dockerfile

@@ -4,7 +4,7 @@ FROM intel/oneapi-basekit:$ONEAPI_VERSION as build
 
 ARG LLAMA_SYCL_F16=OFF
 RUN apt-get update && \
-    apt-get install -y git
+    apt-get install -y git libcurl4-openssl-dev
 
 WORKDIR /app
 
@@ -16,11 +16,14 @@ RUN mkdir build && \
         echo "LLAMA_SYCL_F16 is set" && \
         export OPT_SYCL_F16="-DLLAMA_SYCL_F16=ON"; \
     fi && \
-    cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx ${OPT_SYCL_F16} && \
+    cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_CURL=ON ${OPT_SYCL_F16} && \
     cmake --build . --config Release --target server
 
 FROM intel/oneapi-basekit:$ONEAPI_VERSION as runtime
 
+RUN apt-get update && \
+    apt-get install -y libcurl4-openssl-dev
+
 COPY --from=build /app/build/bin/server /server
 
 ENV LC_ALL=C.utf8

{llama_cpp_python-0.2.59 → llama_cpp_python-0.2.60}/vendor/llama.cpp/.devops/server-rocm.Dockerfile

@@ -40,6 +40,11 @@ ENV LLAMA_HIPBLAS=1
 ENV CC=/opt/rocm/llvm/bin/clang
 ENV CXX=/opt/rocm/llvm/bin/clang++
 
+# Enable cURL
+ENV LLAMA_CURL=1
+RUN apt-get update && \
+    apt-get install -y libcurl4-openssl-dev
+
 RUN make
 
 ENTRYPOINT [ "/app/server" ]