Python: experimental py3.13 support

.github/workflows/python-build.yml

@@ -21,6 +21,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ matrix.python-version }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Check version
         run: |
           echo "Building and uploading Python package version: ${{ github.event.release.tag_name }}"

.github/workflows/python-integration-tests.yml

@@ -120,6 +120,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ matrix.python-version }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Install dependencies
         run: |
           uv sync --all-extras --dev
@@ -257,6 +258,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ matrix.python-version }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Install dependencies
         run: |
           uv sync --all-extras --dev
@@ -321,6 +323,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ matrix.python-version }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Install dependencies
         run: |
           uv sync --all-extras --dev
@@ -379,6 +382,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ matrix.python-version }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Install dependencies
         run: |
           uv sync --all-extras --dev

.github/workflows/python-lint.yml

@@ -30,6 +30,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ matrix.python-version }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Install the project
         run: uv sync --all-extras --dev
       - uses: pre-commit/action@v3.0.1

.github/workflows/python-test-coverage.yml

@@ -32,6 +32,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ env.UV_PYTHON }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Install the project
         run: uv sync --all-extras --dev
       - name: Test with pytest

.github/workflows/python-unit-tests.yml

@@ -20,18 +20,10 @@ jobs:
         python-version: ["3.10", "3.11", "3.12"]
         os: [ubuntu-latest, windows-latest, macos-latest]
         experimental: [false]
-        # include:
-        #   - python-version: "3.13"
-        #     os: "ubuntu-latest"
-        #     experimental: true
-        #   - python-version: "3.13t"
-        #     os: "ubuntu-latest"
-        #     experimental: true
-        #     gil: 0
-        #   - python-version: "3.13t"
-        #     os: "ubuntu-latest"
-        #     experimental: true
-        #     gil: 1
+        include:
+          - python-version: "3.13"
+            os: "ubuntu-latest"
+            experimental: true
     env:
       UV_PYTHON: ${{ matrix.python-version }}
     permissions:
@@ -47,6 +39,7 @@ jobs:
           version: "0.5.x"
           enable-cache: true
           cache-suffix: ${{ runner.os }}-${{ matrix.python-version }}
+          cache-dependency-glob: "**/uv.lock"
       - name: Install the project
         run: uv sync --all-extras --dev -U --prerelease=if-necessary-or-explicit
       - name: Test with pytest

python/.pre-commit-config.yaml

@@ -32,14 +32,14 @@ repos:
     - id: pyupgrade
       args: [--py310-plus]
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.7.4
+    rev: v0.9.6
     hooks:
       - id: ruff
         args: [ --fix, --exit-non-zero-on-fix ]
       - id: ruff-format
   - repo: https://github.com/astral-sh/uv-pre-commit
     # uv version.
-    rev: 0.5.21
+    rev: 0.5.30
     hooks:
       # Update the uv lockfile
       - id: uv-lock

python/Makefile

@@ -32,7 +32,7 @@ help:
 	echo ""
 	echo -e "\033[1mVARIABLES:\033[0m"
 	echo "  PYTHON_VERSION       - Python version to use. Default is 3.10"
-	echo "                        By default, 3.10, 3.11, and 3.12 are installed as well."
+	echo "                        By default, 3.10, 3.11, 3.12 and 3.13 are installed as well."
 
 ##############################
 #  INSTALL
@@ -68,8 +68,8 @@ endif
 ##############################
 .ONESHELL:
 install-python:
-	echo "Installing python 3.10, 3.11, 3.12"
-	uv python install 3.10 3.11 3.12
+	echo "Installing python versions"
+	uv python install 3.10 3.11 3.12 3.13
 
 ##############################
 #  INSTALL-PRE-COMMIT
@@ -87,7 +87,7 @@ install-sk:
 	echo "Creating and activating venv for python $(PYTHON_VERSION)"
 	uv venv --python $(PYTHON_VERSION)
 	echo "Installing Semantic Kernel and all dependencies"
-	uv sync --all-extras --dev
+	uv sync --all-extras --dev --prerelease=if-necessary-or-explicit
 
 ##############################
 #  CLEAN

python/pyproject.toml

@@ -45,6 +45,7 @@ dependencies = [
     "pybars4 ~= 0.9",
     "jinja2 ~= 3.1",
     "nest-asyncio ~= 1.6",
+    "scipy>=1.15.1",
 ]
 
 ### Optional dependencies
@@ -62,7 +63,7 @@ chroma = [
 ]
 google = [
     "google-cloud-aiplatform == 1.79.0",
-    "google-generativeai == 0.7"
+    "google-generativeai ~= 0.8"
 ]
 hugging_face = [
     "transformers[torch] ~= 4.28",
@@ -87,7 +88,7 @@ ollama = [
     "ollama ~= 0.4"
 ]
 onnx = [
-    "onnxruntime-genai ~= 0.5"
+    "onnxruntime-genai ~= 0.5; python_version < '3.13'"
 ]
 anthropic = [
     "anthropic ~= 0.32"
@@ -107,11 +108,11 @@ redis = [
     "redisvl >= 0.3.6",
 ]
 usearch = [
-    "usearch ~= 2.9",
+    "usearch ~= 2.16",
     "pyarrow >= 12.0,< 20.0"
 ]
 weaviate = [
-    "weaviate-client>=4.7,<5.0",
+    "weaviate-client>=4.10,<5.0",
 ]
 pandas = [
     "pandas ~= 2.2"
@@ -139,7 +140,8 @@ dev-dependencies = [
     "snoop ~= 0.4",
     "mypy >= 1.10",
     "types-PyYAML ~= 6.0.12.20240311",
-    "ruff ~= 0.7",
+    "ruff ~= 0.9",
+    "pyright>=1.1.392.post0",
 ]
 environments = [
     "sys_platform == 'darwin'",

python/pyrightconfig.json

@@ -0,0 +1,27 @@
+{
+    "reportMissingImports": false,
+    "include": [
+        "semantic_kernel/**/*.py"
+    ],
+    "pythonVersion": "3.10",
+    "exclude": [
+        "semantic_kernel/memory/*",
+        "semantic_kernel/planners/*",
+        "semantic_kernel/connectors/memory/astradb/*",
+        "semantic_kernel/connectors/memory/azure_ai_search/*",
+        "semantic_kernel/connectors/memory/azure_cognitive_search/*",
+        "semantic_kernel/connectors/memory/azure_cosmosdb/*",
+        "semantic_kernel/connectors/memory/azure_cosmosdb_no_sql/*",
+        "semantic_kernel/connectors/memory/chroma/*",
+        "semantic_kernel/connectors/memory/milvus/*",
+        "semantic_kernel/connectors/memory/mongodb_atlas/*",
+        "semantic_kernel/connectors/memory/pinecone/*",
+        "semantic_kernel/connectors/memory/postgres/*",
+        "semantic_kernel/connectors/memory/qdrant/*",
+        "semantic_kernel/connectors/memory/redis/*",
+        "semantic_kernel/connectors/memory/usearch/*",
+        "semantic_kernel/connectors/memory/weaviate/*"
+    ],
+    "venvPath": ".",
+    "venv": ".venv"
+}

python/samples/demos/document_generator/custom_selection_strategy.py

@@ -17,6 +17,8 @@
     from semantic_kernel.agents import Agent
     from semantic_kernel.contents.chat_message_content import ChatMessageContent
 
+NEWLINE = "\n"
+
 
 @experimental_class
 class CustomSelectionStrategy(SelectionStrategy):
@@ -80,7 +82,7 @@ def get_system_message(self, agents: list["Agent"]) -> str:
 Initially, the chat history may be empty.
 
 Here are the agents with their indices, names, and descriptions:
-{"\n".join(f"[{index}] {agent.name}:\n{agent.description}" for index, agent in enumerate(agents))}
+{NEWLINE.join(f"[{index}] {agent.name}:{NEWLINE}{agent.description}" for index, agent in enumerate(agents))}
 
 Your task is to select the next agent based on the conversation history.
 

python/samples/demos/document_generator/custom_termination_strategy.py

@@ -20,6 +20,8 @@
 TERMINATE_TRUE_KEYWORD = "yes"
 TERMINATE_FALSE_KEYWORD = "no"
 
+NEWLINE = "\n"
+
 
 class CustomTerminationStrategy(TerminationStrategy):
     NUM_OF_RETRIES: ClassVar[int] = 3
@@ -82,7 +84,7 @@ def get_system_message(self) -> str:
 The chat history may start empty as no agents have spoken yet.
 
 Here are the agents with their indices, names, and descriptions:
-{"\n".join(f"[{index}] {agent.name}:\n{agent.description}" for index, agent in enumerate(self.agents))}
+{NEWLINE.join(f"[{index}] {agent.name}:{NEWLINE}{agent.description}" for index, agent in enumerate(self.agents))}
 
 Your task is NOT to continue the conversation. Determine if the latest content is approved by all agents.
 If approved, say "{TERMINATE_TRUE_KEYWORD}". Otherwise, say "{TERMINATE_FALSE_KEYWORD}".

python/samples/demos/guided_conversations/guided_conversation/plugins/agenda.py

@@ -152,7 +152,7 @@ def get_agenda_for_prompt(self) -> str:
             return "None"
         agenda_str = "\n".join(
             [
-                f"{i+1}. [{format_resource(item['resource'], ResourceConstraintUnit.TURNS)}] {item['title']}"
+                f"{i + 1}. [{format_resource(item['resource'], ResourceConstraintUnit.TURNS)}] {item['title']}"
                 for i, item in enumerate(agenda_items)
             ]
         )

python/samples/demos/guided_conversations/guided_conversation/utils/resources.py

@@ -205,7 +205,7 @@ def get_resource_instructions(self) -> tuple[str, str]:
             resource_instructions = ""
 
         if self.resource_constraint.mode == ResourceConstraintMode.EXACT:
-            exact_mode_instructions = f"""There {'are' if is_plural_remaining else 'is'} {formatted_remaining_resource} remaining (including this one) - the conversation will automatically terminate when 0 turns are left. \
+            exact_mode_instructions = f"""There {"are" if is_plural_remaining else "is"} {formatted_remaining_resource} remaining (including this one) - the conversation will automatically terminate when 0 turns are left. \
 You should continue the conversation until it is automatically terminated. This means you should NOT preemptively end the conversation, \
 either explicitly (by selecting the "End conversation" action) or implicitly (e.g. by telling the user that you have all required information and they should wait for the next step). \
 Your goal is not to maximize efficiency (i.e. complete the artifact as quickly as possible then end the conversation), but rather to make the best use of ALL remaining turns available to you"""

python/semantic_kernel/connectors/ai/google/google_ai/services/google_ai_chat_completion.py

@@ -127,16 +127,18 @@ async def _inner_get_chat_message_contents(
         assert isinstance(settings, GoogleAIChatPromptExecutionSettings)  # nosec
 
         genai.configure(api_key=self.service_settings.api_key.get_secret_value())
+        if not self.service_settings.gemini_model_id:
+            raise ServiceInitializationError("The Google AI Gemini model ID is required.")
         model = GenerativeModel(
-            self.service_settings.gemini_model_id,
+            model_name=self.service_settings.gemini_model_id,
             system_instruction=filter_system_message(chat_history),
         )
 
         response: AsyncGenerateContentResponse = await model.generate_content_async(
             contents=self._prepare_chat_history_for_request(chat_history),
             generation_config=GenerationConfig(**settings.prepare_settings_dict()),
             tools=settings.tools,
-            tool_config=settings.tool_config,
+            tool_config=settings.tool_config,  # type: ignore
         )
 
         return [self._create_chat_message_content(response, candidate) for candidate in response.candidates]
@@ -154,16 +156,18 @@ async def _inner_get_streaming_chat_message_contents(
         assert isinstance(settings, GoogleAIChatPromptExecutionSettings)  # nosec
 
         genai.configure(api_key=self.service_settings.api_key.get_secret_value())
+        if not self.service_settings.gemini_model_id:
+            raise ServiceInitializationError("The Google AI Gemini model ID is required.")
         model = GenerativeModel(
-            self.service_settings.gemini_model_id,
+            model_name=self.service_settings.gemini_model_id,
             system_instruction=filter_system_message(chat_history),
         )
 
         response: AsyncGenerateContentResponse = await model.generate_content_async(
             contents=self._prepare_chat_history_for_request(chat_history),
             generation_config=GenerationConfig(**settings.prepare_settings_dict()),
             tools=settings.tools,
-            tool_config=settings.tool_config,
+            tool_config=settings.tool_config,  # type: ignore
             stream=True,
         )
 

python/semantic_kernel/connectors/ai/google/google_ai/services/google_ai_text_completion.py

@@ -99,8 +99,10 @@ async def _inner_get_text_contents(
         assert isinstance(settings, GoogleAITextPromptExecutionSettings)  # nosec
 
         genai.configure(api_key=self.service_settings.api_key.get_secret_value())
+        if not self.service_settings.gemini_model_id:
+            raise ServiceInitializationError("The Google AI Gemini model ID is required.")
         model = GenerativeModel(
-            self.service_settings.gemini_model_id,
+            model_name=self.service_settings.gemini_model_id,
         )
 
         response: AsyncGenerateContentResponse = await model.generate_content_async(
@@ -122,8 +124,10 @@ async def _inner_get_streaming_text_contents(
         assert isinstance(settings, GoogleAITextPromptExecutionSettings)  # nosec
 
         genai.configure(api_key=self.service_settings.api_key.get_secret_value())
+        if not self.service_settings.gemini_model_id:
+            raise ServiceInitializationError("The Google AI Gemini model ID is required.")
         model = GenerativeModel(
-            self.service_settings.gemini_model_id,
+            model_name=self.service_settings.gemini_model_id,
         )
 
         response: AsyncGenerateContentResponse = await model.generate_content_async(

python/semantic_kernel/connectors/ai/google/google_ai/services/google_ai_text_embedding.py

@@ -93,7 +93,9 @@ async def generate_raw_embeddings(
         assert isinstance(settings, GoogleAIEmbeddingPromptExecutionSettings)  # nosec
 
         genai.configure(api_key=self.service_settings.api_key.get_secret_value())
-        response: BatchEmbeddingDict = await genai.embed_content_async(
+        if not self.service_settings.embedding_model_id:
+            raise ServiceInitializationError("The Google AI embedding model ID is required.")
+        response: BatchEmbeddingDict = await genai.embed_content_async(  # type: ignore
             model=self.service_settings.embedding_model_id,
             content=texts,
             **settings.prepare_settings_dict(),

python/semantic_kernel/connectors/ai/hugging_face/hf_prompt_execution_settings.py

@@ -2,10 +2,16 @@
 
 from typing import Any
 
-from transformers import GenerationConfig
-
 from semantic_kernel.connectors.ai.prompt_execution_settings import PromptExecutionSettings
 
+try:
+    from transformers import GenerationConfig
+
+    ready = True
+except ImportError:
+    GenerationConfig = Any
+    ready = False
+
 
 class HuggingFacePromptExecutionSettings(PromptExecutionSettings):
     """Hugging Face prompt execution settings."""
@@ -19,8 +25,10 @@ class HuggingFacePromptExecutionSettings(PromptExecutionSettings):
     temperature: float = 1.0
     top_p: float = 1.0
 
-    def get_generation_config(self) -> GenerationConfig:
+    def get_generation_config(self) -> "GenerationConfig":
         """Get the generation config."""
+        if not ready:
+            raise ImportError("transformers is not installed.")
         return GenerationConfig(
             **self.model_dump(
                 include={"max_new_tokens", "pad_token_id", "eos_token_id", "temperature", "top_p"},

python/semantic_kernel/connectors/ai/hugging_face/services/hf_text_completion.py

@@ -11,6 +11,7 @@
 else:
     from typing_extensions import override  # pragma: no cover
 
+
 import torch
 from transformers import AutoTokenizer, TextIteratorStreamer, pipeline
 

python/semantic_kernel/connectors/ai/hugging_face/services/hf_text_embedding.py

@@ -22,6 +22,7 @@
 
     from semantic_kernel.connectors.ai.prompt_execution_settings import PromptExecutionSettings
 
+
 logger: logging.Logger = logging.getLogger(__name__)