Python: Move ollama samples to samples getting started dir

python/packages/core/agent_framework/ollama/__init__.py

@@ -0,0 +1,23 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import importlib
+from typing import Any
+
+IMPORT_PATH = "agent_framework_ollama"
+PACKAGE_NAME = "agent-framework-ollama"
+_IMPORTS = ["__version__", "OllamaChatClient", "OllamaSettings"]
+
+
+def __getattr__(name: str) -> Any:
+    if name in _IMPORTS:
+        try:
+            return getattr(importlib.import_module(IMPORT_PATH), name)
+        except ModuleNotFoundError as exc:
+            raise ModuleNotFoundError(
+                f"The '{PACKAGE_NAME}' package is not installed, please do `pip install {PACKAGE_NAME}`"
+            ) from exc
+    raise AttributeError(f"Module {IMPORT_PATH} has no attribute {name}.")
+
+
+def __dir__() -> list[str]:
+    return _IMPORTS

python/packages/core/agent_framework/ollama/__init__.pyi

@@ -0,0 +1,13 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+from agent_framework_ollama import (
+    OllamaChatClient,
+    OllamaSettings,
+    __version__,
+)
+
+__all__ = [
+    "OllamaChatClient",
+    "OllamaSettings",
+    "__version__",
+]

python/packages/core/pyproject.toml

@@ -52,6 +52,7 @@ all = [
     "agent-framework-devui",
     "agent-framework-lab",
     "agent-framework-mem0",
+    "agent-framework-ollama",
     "agent-framework-purview",
     "agent-framework-redis",
 ]

python/samples/README.md

@@ -99,11 +99,15 @@ This directory contains samples demonstrating the capabilities of Microsoft Agen
 
 ### Ollama
 
+The recommended way to use Ollama is via the native `OllamaChatClient` from the `agent-framework-ollama` package.
+
 | File | Description |
 |------|-------------|
-| [`getting_started/agents/ollama/ollama_with_openai_chat_client.py`](./getting_started/agents/ollama/ollama_with_openai_chat_client.py) | Ollama with OpenAI Chat Client Example |
-| [`packages/ollama/getting_started/ollama_agent_basic.py`](../packages/ollama/getting_started/ollama_agent_basic.py) | (Experimental) Ollama Agent with native Ollama Chat Client |
-| [`packages/ollama/getting_started/ollama_agent_reasoning.py`](../packages/ollama/getting_started/ollama_agent_reasoning.py) | (Experimental) Ollama Reasoning Agent with native Ollama Chat Client |
+| [`getting_started/agents/ollama/ollama_agent_basic.py`](./getting_started/agents/ollama/ollama_agent_basic.py) | Basic Ollama Agent with native Ollama Chat Client |
+| [`getting_started/agents/ollama/ollama_agent_reasoning.py`](./getting_started/agents/ollama/ollama_agent_reasoning.py) | Ollama Agent with reasoning capabilities |
+| [`getting_started/agents/ollama/ollama_chat_client.py`](./getting_started/agents/ollama/ollama_chat_client.py) | Direct usage of Ollama Chat Client |
+| [`getting_started/agents/ollama/ollama_chat_multimodal.py`](./getting_started/agents/ollama/ollama_chat_multimodal.py) | Ollama Chat Client with multimodal (image) input |
+| [`getting_started/agents/ollama/ollama_with_openai_chat_client.py`](./getting_started/agents/ollama/ollama_with_openai_chat_client.py) | Alternative: Ollama via OpenAI Chat Client |
 
 ### OpenAI
 
@@ -149,7 +153,6 @@ This directory contains samples demonstrating the capabilities of Microsoft Agen
 | [`getting_started/chat_client/openai_assistants_client.py`](./getting_started/chat_client/openai_assistants_client.py) | OpenAI Assistants Client Direct Usage Example |
 | [`getting_started/chat_client/openai_chat_client.py`](./getting_started/chat_client/openai_chat_client.py) | OpenAI Chat Client Direct Usage Example |
 | [`getting_started/chat_client/openai_responses_client.py`](./getting_started/chat_client/openai_responses_client.py) | OpenAI Responses Client Direct Usage Example |
-| [`packages/ollama/getting_started/ollama_chat_client.py`](../packages/ollama/getting_started/ollama_chat_client.py) | (Experimental) Ollama Chat Client with native Ollama Chat Client |
 
 
 ## Context Providers
@@ -225,7 +228,6 @@ This directory contains samples demonstrating the capabilities of Microsoft Agen
 | [`getting_started/multimodal_input/azure_chat_multimodal.py`](./getting_started/multimodal_input/azure_chat_multimodal.py) | Azure OpenAI Chat with multimodal (image) input example |
 | [`getting_started/multimodal_input/azure_responses_multimodal.py`](./getting_started/multimodal_input/azure_responses_multimodal.py) | Azure OpenAI Responses with multimodal (image) input example |
 | [`getting_started/multimodal_input/openai_chat_multimodal.py`](./getting_started/multimodal_input/openai_chat_multimodal.py) | OpenAI Chat with multimodal (image) input example |
-| [`packages/ollama/getting_started/ollama_chat_multimodal.py`](../packages/ollama/getting_started/ollama_chat_multimodal.py) | (Experimental) Ollama Chat  with multimodal native Ollama Chat Client |
 
 
 ## Azure Functions

python/samples/getting_started/agents/ollama/README.md

@@ -8,20 +8,41 @@ This folder contains examples demonstrating how to use Ollama models with the Ag
 2. **Start Ollama**: Ensure Ollama is running on your local machine
 3. **Pull a model**: Run `ollama pull mistral` (or any other model you prefer)
    - For function calling examples, use models that support tool calling like `mistral` or `qwen2.5`
-   - For reasoning examples, use models that support reasoning like `qwen2.5:8b`
+   - For reasoning examples, use models that support reasoning like `qwen3:8b`
+   - For multimodal examples, use models like `gemma3:4b`
 
-> **Note**: Not all models support all features. Function calling and reasoning capabilities depend on the specific model you're using.
+> **Note**: Not all models support all features. Function calling, reasoning, and multimodal capabilities depend on the specific model you're using.
+
+## Recommended Approach
+
+The recommended way to use Ollama with Agent Framework is via the native `OllamaChatClient` from the `agent-framework-ollama` package. This provides full support for Ollama-specific features like reasoning mode.
+
+Alternatively, you can use the `OpenAIChatClient` configured to point to your local Ollama server, which may be useful if you're already familiar with the OpenAI client interface.
 
 ## Examples
 
 | File | Description |
 |------|-------------|
-| [`ollama_with_openai_chat_client.py`](ollama_with_openai_chat_client.py) | Demonstrates how to configure OpenAI Chat Client to use local Ollama models. Shows both streaming and non-streaming responses with tool calling capabilities. |
+| [`ollama_agent_basic.py`](ollama_agent_basic.py) | Basic Ollama agent with tool calling using native Ollama Chat Client. Shows both streaming and non-streaming responses. |
+| [`ollama_agent_reasoning.py`](ollama_agent_reasoning.py) | Ollama agent with reasoning capabilities using native Ollama Chat Client. Shows how to enable thinking/reasoning mode. |
+| [`ollama_chat_client.py`](ollama_chat_client.py) | Direct usage of the native Ollama Chat Client with tool calling. |
+| [`ollama_chat_multimodal.py`](ollama_chat_multimodal.py) | Ollama Chat Client with multimodal (image) input capabilities. |
+| [`ollama_with_openai_chat_client.py`](ollama_with_openai_chat_client.py) | Alternative approach using OpenAI Chat Client configured to use local Ollama models. |
 
 ## Configuration
 
 The examples use environment variables for configuration. Set the appropriate variables based on which example you're running:
 
+### For Native Ollama Examples
+
+Set the following environment variables:
+
+- `OLLAMA_HOST`: The base URL for your Ollama server (optional, defaults to `http://localhost:11434`)
+  - Example: `export OLLAMA_HOST="http://localhost:11434"`
+
+- `OLLAMA_CHAT_MODEL_ID`: The model name to use
+  - Example: `export OLLAMA_CHAT_MODEL_ID="qwen2.5:8b"`
+  - Must be a model you have pulled with Ollama
 
 ### For OpenAI Client with Ollama (`ollama_with_openai_chat_client.py`)
 

python/packages/ollama/getting_started/ollama_agent_basic.py → python/samples/getting_started/agents/ollama/ollama_agent_basic.py

@@ -3,7 +3,7 @@
 import asyncio
 from datetime import datetime
 
-from agent_framework_ollama import OllamaChatClient
+from agent_framework.ollama import OllamaChatClient
 
 """
 Ollama Agent Basic Example

python/packages/ollama/getting_started/ollama_agent_reasoning.py → python/samples/getting_started/agents/ollama/ollama_agent_reasoning.py

@@ -3,8 +3,7 @@
 import asyncio
 
 from agent_framework import TextReasoningContent
-
-from agent_framework_ollama import OllamaChatClient
+from agent_framework.ollama import OllamaChatClient
 
 """
 Ollama Agent Reasoning Example

python/packages/ollama/getting_started/ollama_chat_client.py → python/samples/getting_started/agents/ollama/ollama_chat_client.py

@@ -3,12 +3,19 @@
 import asyncio
 from datetime import datetime
 
-from agent_framework_ollama import OllamaChatClient
+from agent_framework.ollama import OllamaChatClient
 
-# Ensure to install Ollama and have a model running locally before running the sample
-# Not all Models support function calling, to test function calling try llama3.2
-# Set the model to use via the OLLAMA_CHAT_MODEL_ID environment variable or modify the code below.
-# https://ollama.com/
+"""
+Ollama Chat Client Example
+
+This sample demonstrates using the native Ollama Chat Client directly.
+
+Ensure to install Ollama and have a model running locally before running the sample.
+Not all Models support function calling, to test function calling try llama3.2
+Set the model to use via the OLLAMA_CHAT_MODEL_ID environment variable or modify the code below.
+https://ollama.com/
+
+"""
 
 
 def get_time():

python/packages/ollama/getting_started/ollama_chat_multimodal.py → python/samples/getting_started/agents/ollama/ollama_chat_multimodal.py

@@ -3,8 +3,7 @@
 import asyncio
 
 from agent_framework import ChatMessage, DataContent, Role, TextContent
-
-from agent_framework_ollama import OllamaChatClient
+from agent_framework.ollama import OllamaChatClient
 
 """
 Ollama Agent Multimodal Example