not-again-ai 0.13.0__tar.gz → 0.14.0__tar.gz

{not_again_ai-0.13.0 → not_again_ai-0.14.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: not-again-ai
-Version: 0.13.0
+Version: 0.14.0
 Summary: Designed to once and for all collect all the little things that come up over and over again in AI projects and put them in one place.
 Home-page: https://github.com/DaveCoDev/not-again-ai
 License: MIT
@@ -17,24 +17,26 @@ Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Typing :: Typed
+Provides-Extra: data
 Provides-Extra: llm
 Provides-Extra: local-llm
 Provides-Extra: statistics
 Provides-Extra: viz
-Requires-Dist: azure-ai-inference (==1.0.0b4) ; extra == "llm"
-Requires-Dist: azure-identity (>=1.18,<2.0) ; extra == "llm"
+Requires-Dist: azure-ai-inference (==1.0.0b5) ; extra == "llm"
+Requires-Dist: azure-identity (>=1.19,<2.0) ; extra == "llm"
 Requires-Dist: jinja2 (>=3.1,<4.0) ; extra == "local-llm"
-Requires-Dist: loguru (==0.7.2)
+Requires-Dist: loguru (>=0.7,<0.8)
 Requires-Dist: numpy (>=2.1,<3.0) ; extra == "statistics" or extra == "viz"
 Requires-Dist: ollama (>=0.3,<0.4) ; extra == "local-llm"
-Requires-Dist: openai (>=1.51,<2.0) ; extra == "llm"
+Requires-Dist: openai (>=1.52,<2.0) ; extra == "llm"
 Requires-Dist: pandas (>=2.2,<3.0) ; extra == "viz"
-Requires-Dist: pydantic (>=2.9,<3.0) ; extra == "llm"
+Requires-Dist: pydantic (>=2.9,<3.0)
+Requires-Dist: pytest-playwright (>=0.5,<0.6) ; extra == "data"
 Requires-Dist: python-liquid (>=1.12,<2.0) ; extra == "llm"
 Requires-Dist: scikit-learn (>=1.5,<2.0) ; extra == "statistics"
 Requires-Dist: scipy (>=1.14,<2.0) ; extra == "statistics"
 Requires-Dist: seaborn (>=0.13,<0.14) ; extra == "viz"
-Requires-Dist: tiktoken (>=0.7,<0.8) ; extra == "llm"
+Requires-Dist: tiktoken (>=0.8,<0.9) ; extra == "llm"
 Requires-Dist: transformers (>=4.45,<5.0) ; extra == "local-llm"
 Project-URL: Documentation, https://github.com/DaveCoDev/not-again-ai
 Project-URL: Repository, https://github.com/DaveCoDev/not-again-ai
@@ -72,34 +74,53 @@ $ pip install not_again_ai[llm,local_llm,statistics,viz]
 Note that local LLM requires separate installations and will not work out of the box due to how hardware dependent it is. Be sure to check the [notebooks](notebooks/local_llm/) for more details.
 
 The package is split into subpackages, so you can install only the parts you need.
-* **Base only**: `pip install not_again_ai`
-* **LLM**: `pip install not_again_ai[llm]`
-    1. OpenAI API
-        1. Go to https://platform.openai.com/settings/profile?tab=api-keys to get your API key.
-        1. (Optional) Set the `OPENAI_API_KEY` and the `OPENAI_ORG_ID` environment variables.
-    1. Azure OpenAI (AOAI)
-       1. Using AOAI requires using Entra ID authentication. See https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/managed-identity for how to set this up for your AOAI deployment.
-          1. Requires the correct role assigned to your user account and being signed into the Azure CLI.
-       1. (Optional) Set the `AZURE_OPENAI_ENDPOINT` environment variable.
-    1. GitHub Models
-       1. Get a Personal Access Token from https://github.com/settings/tokens and set the `GITHUB_TOKEN` environment variable. The token does not need any permissions.
-       1. Check the [Github Marketplace](https://github.com/marketplace/models) to see which models are available.
-* **Local LLM**: `pip install not_again_ai[llm,local_llm]`
-    1. Some HuggingFace transformers tokenizers are gated behind access requests. If you wish to use these, you will need to request access from HuggingFace on the model card.
-       1. Then set the `HF_TOKEN` environment variable to your HuggingFace API token which can be found here: https://huggingface.co/settings/tokens
-    2. If you wish to use Ollama:
-        1. Follow the instructions at https://github.com/ollama/ollama to install Ollama for your system. 
-        2. (Optional) [Add Ollama as a startup service (recommended)](https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended)
-        3. (Optional) To make the Ollama service accessible on your local network from a Linux server, add the following to the `/etc/systemd/system/ollama.service` file which will make Ollama available at `http://<local_address>:11434`:
-            ```bash
-            [Service]
-            ...
-            Environment="OLLAMA_HOST=0.0.0.0"
-            ```
-        4. It is recommended to always have the latest version of Ollama. To update Ollama check the [docs](https://github.com/ollama/ollama/blob/main/docs/). The command for Linux is: `curl -fsSL https://ollama.com/install.sh | sh`
-    3. HuggingFace transformers and other requirements are hardware dependent so for providers other than Ollama, this only installs some generic dependencies. Check the [notebooks](notebooks/local_llm/) for more details on what is available and how to install it.
-* **Statistics**: `pip install not_again_ai[statistics]`
-* **Visualization**: `pip install not_again_ai[viz]`
+
+### Base
+1. `pip install not_again_ai`
+
+
+### Data
+1. `pip install not_again_ai[data]`
+1. `playwright install` to download the browser binaries.
+
+
+### LLM
+1. `pip install not_again_ai[llm]`
+1. Setup OpenAI API
+   1. Go to https://platform.openai.com/settings/profile?tab=api-keys to get your API key.
+   1. (Optional) Set the `OPENAI_API_KEY` and the `OPENAI_ORG_ID` environment variables.
+1. Setup Azure OpenAI (AOAI)
+   1. Using AOAI requires using Entra ID authentication. See https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/managed-identity for how to set this up for your AOAI deployment.
+      * Requires the correct role assigned to your user account and being signed into the Azure CLI.
+   1. (Optional) Set the `AZURE_OPENAI_ENDPOINT` environment variable.
+1. Setup GitHub Models
+   1. Get a Personal Access Token from https://github.com/settings/tokens and set the `GITHUB_TOKEN` environment variable. The token does not need any permissions.
+   1. Check the [Github Marketplace](https://github.com/marketplace/models) to see which models are available.
+
+
+### Local LLM
+ 1. `pip install not_again_ai[llm,local_llm]`
+ 1. Some HuggingFace transformers tokenizers are gated behind access requests. If you wish to use these, you will need to request access from HuggingFace on the model card.
+    * Then set the `HF_TOKEN` environment variable to your HuggingFace API token which can be found here: https://huggingface.co/settings/tokens
+ 1. If you wish to use Ollama:
+     1. Follow the instructions at https://github.com/ollama/ollama to install Ollama for your system. 
+     1. (Optional) [Add Ollama as a startup service (recommended)](https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended)
+     1. (Optional) To make the Ollama service accessible on your local network from a Linux server, add the following to the `/etc/systemd/system/ollama.service` file which will make Ollama available at `http://<local_address>:11434`:
+         ```bash
+         [Service]
+         ...
+         Environment="OLLAMA_HOST=0.0.0.0"
+         ```
+     1. It is recommended to always have the latest version of Ollama. To update Ollama check the [docs](https://github.com/ollama/ollama/blob/main/docs/). The command for Linux is: `curl -fsSL https://ollama.com/install.sh | sh`
+ 1. HuggingFace transformers and other requirements are hardware dependent so for providers other than Ollama, this only installs some generic dependencies. Check the [notebooks](notebooks/local_llm/) for more details on what is available and how to install it.
+
+
+### Statistics
+1. `pip install not_again_ai[statistics]`
+
+
+### Visualization
+1. `pip install not_again_ai[viz]`
 
 
 # Development Information
@@ -229,10 +250,10 @@ areas of the project that are currently not tested.
 
 pytest and code coverage are configured in [`pyproject.toml`](./pyproject.toml).
 
-To pass arguments to `pytest` through `nox`:
+To run selected tests:
 
 ```bash
-(.venv) $ nox -s test -- -k invalid_factorial
+(.venv) $ nox -s test -- -k "test_web"
 ```
 
 ## Code Style Checking

{not_again_ai-0.13.0 → not_again_ai-0.14.0}/README.md

@@ -30,34 +30,53 @@ $ pip install not_again_ai[llm,local_llm,statistics,viz]
 Note that local LLM requires separate installations and will not work out of the box due to how hardware dependent it is. Be sure to check the [notebooks](notebooks/local_llm/) for more details.
 
 The package is split into subpackages, so you can install only the parts you need.
-* **Base only**: `pip install not_again_ai`
-* **LLM**: `pip install not_again_ai[llm]`
-    1. OpenAI API
-        1. Go to https://platform.openai.com/settings/profile?tab=api-keys to get your API key.
-        1. (Optional) Set the `OPENAI_API_KEY` and the `OPENAI_ORG_ID` environment variables.
-    1. Azure OpenAI (AOAI)
-       1. Using AOAI requires using Entra ID authentication. See https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/managed-identity for how to set this up for your AOAI deployment.
-          1. Requires the correct role assigned to your user account and being signed into the Azure CLI.
-       1. (Optional) Set the `AZURE_OPENAI_ENDPOINT` environment variable.
-    1. GitHub Models
-       1. Get a Personal Access Token from https://github.com/settings/tokens and set the `GITHUB_TOKEN` environment variable. The token does not need any permissions.
-       1. Check the [Github Marketplace](https://github.com/marketplace/models) to see which models are available.
-* **Local LLM**: `pip install not_again_ai[llm,local_llm]`
-    1. Some HuggingFace transformers tokenizers are gated behind access requests. If you wish to use these, you will need to request access from HuggingFace on the model card.
-       1. Then set the `HF_TOKEN` environment variable to your HuggingFace API token which can be found here: https://huggingface.co/settings/tokens
-    2. If you wish to use Ollama:
-        1. Follow the instructions at https://github.com/ollama/ollama to install Ollama for your system. 
-        2. (Optional) [Add Ollama as a startup service (recommended)](https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended)
-        3. (Optional) To make the Ollama service accessible on your local network from a Linux server, add the following to the `/etc/systemd/system/ollama.service` file which will make Ollama available at `http://<local_address>:11434`:
-            ```bash
-            [Service]
-            ...
-            Environment="OLLAMA_HOST=0.0.0.0"
-            ```
-        4. It is recommended to always have the latest version of Ollama. To update Ollama check the [docs](https://github.com/ollama/ollama/blob/main/docs/). The command for Linux is: `curl -fsSL https://ollama.com/install.sh | sh`
-    3. HuggingFace transformers and other requirements are hardware dependent so for providers other than Ollama, this only installs some generic dependencies. Check the [notebooks](notebooks/local_llm/) for more details on what is available and how to install it.
-* **Statistics**: `pip install not_again_ai[statistics]`
-* **Visualization**: `pip install not_again_ai[viz]`
+
+### Base
+1. `pip install not_again_ai`
+
+
+### Data
+1. `pip install not_again_ai[data]`
+1. `playwright install` to download the browser binaries.
+
+
+### LLM
+1. `pip install not_again_ai[llm]`
+1. Setup OpenAI API
+   1. Go to https://platform.openai.com/settings/profile?tab=api-keys to get your API key.
+   1. (Optional) Set the `OPENAI_API_KEY` and the `OPENAI_ORG_ID` environment variables.
+1. Setup Azure OpenAI (AOAI)
+   1. Using AOAI requires using Entra ID authentication. See https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/managed-identity for how to set this up for your AOAI deployment.
+      * Requires the correct role assigned to your user account and being signed into the Azure CLI.
+   1. (Optional) Set the `AZURE_OPENAI_ENDPOINT` environment variable.
+1. Setup GitHub Models
+   1. Get a Personal Access Token from https://github.com/settings/tokens and set the `GITHUB_TOKEN` environment variable. The token does not need any permissions.
+   1. Check the [Github Marketplace](https://github.com/marketplace/models) to see which models are available.
+
+
+### Local LLM
+ 1. `pip install not_again_ai[llm,local_llm]`
+ 1. Some HuggingFace transformers tokenizers are gated behind access requests. If you wish to use these, you will need to request access from HuggingFace on the model card.
+    * Then set the `HF_TOKEN` environment variable to your HuggingFace API token which can be found here: https://huggingface.co/settings/tokens
+ 1. If you wish to use Ollama:
+     1. Follow the instructions at https://github.com/ollama/ollama to install Ollama for your system. 
+     1. (Optional) [Add Ollama as a startup service (recommended)](https://github.com/ollama/ollama/blob/main/docs/linux.md#adding-ollama-as-a-startup-service-recommended)
+     1. (Optional) To make the Ollama service accessible on your local network from a Linux server, add the following to the `/etc/systemd/system/ollama.service` file which will make Ollama available at `http://<local_address>:11434`:
+         ```bash
+         [Service]
+         ...
+         Environment="OLLAMA_HOST=0.0.0.0"
+         ```
+     1. It is recommended to always have the latest version of Ollama. To update Ollama check the [docs](https://github.com/ollama/ollama/blob/main/docs/). The command for Linux is: `curl -fsSL https://ollama.com/install.sh | sh`
+ 1. HuggingFace transformers and other requirements are hardware dependent so for providers other than Ollama, this only installs some generic dependencies. Check the [notebooks](notebooks/local_llm/) for more details on what is available and how to install it.
+
+
+### Statistics
+1. `pip install not_again_ai[statistics]`
+
+
+### Visualization
+1. `pip install not_again_ai[viz]`
 
 
 # Development Information
@@ -187,10 +206,10 @@ areas of the project that are currently not tested.
 
 pytest and code coverage are configured in [`pyproject.toml`](./pyproject.toml).
 
-To pass arguments to `pytest` through `nox`:
+To run selected tests:
 
 ```bash
-(.venv) $ nox -s test -- -k invalid_factorial
+(.venv) $ nox -s test -- -k "test_web"
 ```
 
 ## Code Style Checking

{not_again_ai-0.13.0 → not_again_ai-0.14.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "not-again-ai"
-version = "0.13.0"
+version = "0.14.0"
 description = "Designed to once and for all collect all the little things that come up over and over again in AI projects and put them in one place."
 authors = ["DaveCoDev <dave.co.dev@gmail.com>"]
 license = "MIT"
@@ -26,26 +26,29 @@ classifiers = [
 # result in an old version being resolved/locked.
 python = "^3.11 || ^3.12"
 
-loguru = { version = "==0.7.2" }
+loguru = { version = "^0.7" }
+pydantic = { version = "^2.9"}
+
 
 # Optional dependencies are defined here, and groupings are defined below.
-azure-ai-inference = { version = "==1.0.0b4", optional = true }
-azure-identity = { version = "^1.18", optional = true }
+azure-ai-inference = { version = "==1.0.0b5", optional = true }
+azure-identity = { version = "^1.19", optional = true }
 jinja2 = { version = "^3.1", optional = true }
 numpy = { version = "^2.1", optional = true }
 ollama = { version = "^0.3", optional = true }
-openai = { version = "^1.51", optional = true }
+openai = { version = "^1.52", optional = true }
 pandas = { version = "^2.2", optional = true }
-pydantic = { version = "^2.9", optional = true }
+pytest-playwright = { version = "^0.5", optional = true }
 python-liquid = { version = "^1.12", optional = true }
 scipy = { version = "^1.14", optional = true }
 scikit-learn = { version = "^1.5", optional = true }
 seaborn = { version = "^0.13", optional = true }
-tiktoken = { version = "^0.7", optional = true }
+tiktoken = { version = "^0.8", optional = true }
 transformers = { version = "^4.45", optional = true }
 
 [tool.poetry.extras]
-llm = ["azure-ai-inference", "azure-identity", "openai", "pydantic", "python-liquid", "tiktoken"]
+data = ["pytest-playwright"]
+llm = ["azure-ai-inference", "azure-identity", "openai", "python-liquid", "tiktoken"]
 local_llm = ["jinja2", "ollama", "transformers"]
 statistics = ["numpy", "scikit-learn", "scipy"]
 viz = ["numpy", "pandas", "seaborn"]

not_again_ai-0.14.0/src/not_again_ai/data/__init__.py

@@ -0,0 +1,7 @@
+import importlib.util
+
+if importlib.util.find_spec("playwright") is None:
+    raise ImportError(
+        "not_again_ai.data requires the 'data' extra to be installed. "
+        "You can install it using 'pip install not_again_ai[data]'."
+    )

not_again_ai-0.14.0/src/not_again_ai/data/web.py

@@ -0,0 +1,56 @@
+from loguru import logger
+from playwright.sync_api import Browser, Playwright, sync_playwright
+
+
+def create_browser(headless: bool = True) -> tuple[Playwright, Browser]:
+    """Creates and returns a new Playwright instance and browser.
+
+    Args:
+        headless (bool, optional): Whether to run the browser in headless mode. Defaults to True.
+
+    Returns:
+        tuple[Playwright, Browser]: A tuple containing the Playwright instance and browser.
+    """
+    pwright = sync_playwright().start()
+    browser = pwright.chromium.launch(
+        headless=headless,
+        chromium_sandbox=False,
+        timeout=15000,
+    )
+    return pwright, browser
+
+
+def get_raw_web_content(url: str, browser: Browser | None = None, headless: bool = True) -> str:
+    """Fetches raw web content from a given URL using Playwright.
+
+    Args:
+        url (str): The URL to fetch content from.
+        browser (Browser | None, optional): An existing browser instance to use. Defaults to None.
+        headless (bool, optional): Whether to run the browser in headless mode. Defaults to True.
+
+    Returns:
+        str: The raw web content.
+    """
+    p = None
+    try:
+        if browser is None:
+            p, browser = create_browser(headless)
+
+        page = browser.new_page(
+            accept_downloads=False,
+            java_script_enabled=True,
+            viewport={"width": 1366, "height": 768},
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/129.0.0.0 Safari/537.3",
+        )
+        page.goto(url)
+        content = page.content()
+        page.close()
+        return content
+    except Exception as e:
+        logger.error(f"Failed to get web content: {e}")
+        return ""
+    finally:
+        if browser:
+            browser.close()
+        if p:
+            p.stop()

{not_again_ai-0.13.0 → not_again_ai-0.14.0}/src/not_again_ai/llm/openai_api/chat_completion.py

@@ -1,3 +1,4 @@
+from collections.abc import Generator
 import contextlib
 import json
 import time
@@ -32,7 +33,7 @@ def chat_completion(
         model (str): ID of the model to use. See the model endpoint compatibility table:
             https://platform.openai.com/docs/models/model-endpoint-compatibility
             for details on which models work with the Chat API.
-        client (OpenAI): An instance of the OpenAI or AzureOpenAI client.
+        client (OpenAI | AzureOpenAI | Any): An instance of the OpenAI or AzureOpenAI client.
             If anything else is provided, we assume that it follows the OpenAI spec and call it by passing kwargs directly.
             For example you can provide something like:
             ```
@@ -198,3 +199,141 @@ def chat_completion(
         del response_data["choices"]
 
     return response_data
+
+
+def chat_completion_stream(
+    messages: list[dict[str, Any]],
+    model: str,
+    client: OpenAI | AzureOpenAI | Any,
+    tools: list[dict[str, Any]] | None = None,
+    tool_choice: str = "auto",
+    max_tokens: int | None = None,
+    temperature: float = 0.7,
+    seed: int | None = None,
+    **kwargs: Any,
+) -> Generator[dict[str, Any], None, None]:
+    """Stream a chat completion from the OpenAI API.
+
+    Args:
+        messages (list[dict[str, Any]]): The messages to send to the model.
+        model (str): The model to use for the chat completion.
+        client (OpenAI | AzureOpenAI | Any): The client to use to send the request.
+            If anything else is provided, we assume that it follows the OpenAI spec and call it by passing kwargs directly.
+            For example you can provide something like:
+            ```
+            def custom_client(**kwargs) -> Generator[dict[str, Any], None, None]:  # type: ignore
+                client = openai_client()
+                completion = client.chat.completions.create(**kwargs)
+                for chunk in completion:
+                    yield chunk.to_dict()
+            ```
+        tools (list[dict[str, Any]], optional):A list of tools the model may call.
+            Use this to provide a list of functions the model may generate JSON inputs for. Defaults to None.
+        tool_choice (str, optional): The tool choice to use. Can be "auto", "required", "none", or a specific function name.
+            Note the function name cannot be any of "auto", "required", or "none". Defaults to "auto".
+        max_tokens (int | None): The maximum number of tokens to generate.
+        temperature (float): The temperature to use for the chat completion.
+        seed (int, optional): If specified, OpenAI will make a best effort to sample deterministically,
+            such that repeated requests with the same `seed` and parameters should return the same result.
+            Does not currently return `system_fingerprint`.
+
+    Returns:
+        Generator[dict[str, Any], None, None]: A generator of chunks of the chat completion.
+        Each chunk is a dictionary with the following keys:
+            role (str): The role of the chunk. Can be "assistant", "tool", or "usage".
+            content (str): The content of the chunk.
+            tool_name (str | None): The name of the tool called by the model.
+            tool_call_id (str | None): The ID of the tool call.
+            completion_tokens (int | None): The number of tokens used by the model to generate the completion.
+            prompt_tokens (int | None): The number of tokens in the messages sent to the model.
+    """
+
+    class ChatCompletionStreamParser:
+        def __init__(self) -> None:
+            # Remembers if we are currently streaming an assistant message or tool call
+            self.last_type: str = ""
+            self.last_tool_name: str | None = None
+            self.last_tool_call_id: str | None = None
+
+        def process_chunk(self, chunk: dict[str, Any]) -> dict[str, Any] | None:
+            """Convert the current chunk into a more digestible format
+            {
+                "role": Literal["assistant", "tool", "usage"],
+                "content": str,
+                "tool_name": str | None,
+                "tool_call_id": str | None,
+                "completion_tokens": int | None,
+                "prompt_tokens": int | None,
+            }
+            """
+            processed_chunk: dict[str, Any] = {}
+            if chunk["choices"]:
+                choice = chunk["choices"][0]
+                # This checks if its just a regular message currently being streamed
+                if choice["delta"].get("role", "") and choice["delta"].get("tool_calls", None) is None:
+                    if choice["delta"]["role"] != self.last_type:
+                        self.last_type = choice["delta"]["role"]
+                        processed_chunk["role"] = self.last_type
+                        if not choice["delta"]["content"]:
+                            processed_chunk["content"] = ""
+                        else:
+                            processed_chunk["content"] = choice["delta"]["content"]
+                    else:
+                        processed_chunk["role"] = self.last_type
+                elif choice["delta"].get("tool_calls", None):
+                    # tool_calls will always be present if the model is calling a tool
+                    tool_call = choice["delta"]["tool_calls"][0]
+                    if tool_call["function"].get("name"):
+                        self.last_type = "tool"
+                        self.last_tool_name = tool_call["function"]["name"]
+                        self.last_tool_call_id = tool_call["id"]
+                    processed_chunk["role"] = "tool"
+                    processed_chunk["content"] = tool_call["function"]["arguments"]
+                    processed_chunk["tool_name"] = self.last_tool_name
+                    processed_chunk["tool_call_id"] = self.last_tool_call_id
+                elif choice["delta"].get("content", ""):
+                    # This is the case after the first regular assistant message
+                    processed_chunk["role"] = self.last_type
+                    processed_chunk["content"] = choice["delta"]["content"]
+            else:
+                if chunk.get("usage"):
+                    processed_chunk["role"] = "usage"
+                    processed_chunk["completion_tokens"] = chunk["usage"]["completion_tokens"]
+                    processed_chunk["prompt_tokens"] = chunk["usage"]["prompt_tokens"]
+                else:
+                    return None
+            return processed_chunk
+
+    kwargs.update(
+        {
+            "messages": messages,
+            "model": model,
+            "max_tokens": max_tokens,
+            "temperature": temperature,
+            "stream": True,
+            "stream_options": {"include_usage": True},
+        }
+    )
+
+    if tools is not None:
+        kwargs["tools"] = tools
+        if tool_choice not in ["none", "auto", "required"]:
+            kwargs["tool_choice"] = {"type": "function", "function": {"name": tool_choice}}
+        else:
+            kwargs["tool_choice"] = tool_choice
+
+    if seed is not None:
+        kwargs["seed"] = seed
+
+    if isinstance(client, OpenAI | AzureOpenAI):
+        response = client.chat.completions.create(**kwargs)
+    else:
+        response = client(**kwargs)
+
+    parser = ChatCompletionStreamParser()
+    for chunk in response:
+        if isinstance(client, OpenAI | AzureOpenAI):
+            chunk = chunk.to_dict()
+        processed_chunk = parser.process_chunk(chunk)
+        if processed_chunk:
+            yield processed_chunk