LLM Docs Nits (#17999)

docs/docs/getting_started/starter_example.md

@@ -21,6 +21,9 @@ export OPENAI_API_KEY=XXXXX
 set OPENAI_API_KEY=XXXXX
 ```
 
+!!! tip
+    If you are using an OpenAI-Compatible API, you can use the `OpenAILike` LLM class. You can find more information in the [OpenAILike LLM](https://docs.llamaindex.ai/en/stable/api_reference/llms/openai_like/#llama_index.llms.openai_like.OpenAILike) integration.
+
 ## Basic Agent Example
 
 Let's start with a simple example using an agent that can perform basic multiplication by calling a tool. Create a file called `starter.py`:

docs/docs/understanding/using_llms/using_llms.md

@@ -47,6 +47,8 @@ messages = [
 chat_response = llm.chat(messages)
 ```
 
+`stream_chat` and `astream_chat` are also available.
+
 ## Specifying models
 
 Many LLM integrations provide more than one model. You can specify a model by passing the `model` parameter to the LLM constructor:
@@ -57,6 +59,60 @@ response = llm.complete("Who is Laurie Voss?")
 print(response)
 ```
 
+## Multi-Modal LLMs
+
+Some LLMs support multi-modal chat messages. This means that you can pass in a mix of text and other modalities (images, audio, video, etc.) and the LLM will handle it.
+
+Currently, LlamaIndex supports text, images, and audio inside ChatMessages using content blocks.
+
+```python
+from llama_index.core.llms import ChatMessage, TextBlock, ImageBlock
+from llama_index.llms.openai import OpenAI
+
+llm = OpenAI(model="gpt-4o")
+
+messages = [
+    ChatMessage(
+        role="user",
+        blocks=[
+            ImageBlock(path="image.png"),
+            TextBlock(text="Describe the image in a few sentences."),
+        ],
+    )
+]
+
+resp = llm.chat(messages)
+print(resp.message.content)
+```
+
+## Tool Calling
+
+Some LLMs (OpenAI, Anthropic, Gemini, Ollama, etc.) support tool calling directly over API calls -- this means tools and functions can be called without specific prompts and parsing mechanisms.
+
+```python
+from llama_index.core.tools import FunctionTool
+from llama_index.llms.openai import OpenAI
+
+
+def generate_song(name: str, artist: str) -> Song:
+    """Generates a song with provided name and artist."""
+    return {"name": name, "artist": artist}
+
+
+tool = FunctionTool.from_defaults(fn=generate_song)
+
+llm = OpenAI(model="gpt-4o")
+response = llm.predict_and_call(
+    [tool],
+    "Pick a random song for me",
+)
+print(str(response))
+```
+
+For more details on even more advanced tool calling, check out the in-depth guide using [OpenAI](../../examples/llm/openai.ipynb). The same approaches work for any LLM that supports tools/functions (e.g. Anthropic, Gemini, Ollama, etc.).
+
+You can learn more about tools and agents in the [tools guide](../../understanding/agent/tools.md).
+
 ## Available LLMs
 
 We support integrations with OpenAI, Anthropic, Mistral, DeepSeek, Hugging Face, and dozens more. Check out our [module guide to LLMs](../../module_guides/models/llms.md) for a full list, including how to run a local model.

llama-index-integrations/llms/llama-index-llms-openai-like/llama_index/llms/openai_like/base.py

@@ -27,13 +27,50 @@ class OpenAILike(OpenAI):
     OpenAILike is a thin wrapper around the OpenAI model that makes it compatible with
     3rd party tools that provide an openai-compatible api.
 
-    Currently, llama_index prevents using custom models with their OpenAI class
-    because they need to be able to infer some metadata from the model name.
-
-    NOTE: You still need to set the OPENAI_BASE_API and OPENAI_API_KEY environment
-    variables or the api_key and api_base constructor arguments.
-    OPENAI_API_KEY/api_key can normally be set to anything in this case,
-    but will depend on the tool you're using.
+    Args:
+        model (str):
+            The model to use for the api.
+        api_base (str):
+            The base url to use for the api.
+            Defaults to "https://api.openai.com/v1".
+        is_chat_model (bool):
+            Whether the model uses the chat or completion endpoint.
+            Defaults to False.
+        is_function_calling_model (bool):
+            Whether the model supports OpenAI function calling/tools over the API.
+            Defaults to False.
+        api_key (str):
+            The api key to use for the api.
+            Set this to some random string if your API does not require an api key.
+        context_window (int):
+            The context window to use for the api. Set this to your model's context window for the best experience.
+            Defaults to 3900.
+        max_tokens (int):
+            The max number of tokens to generate.
+            Defaults to None.
+        temperature (float):
+            The temperature to use for the api.
+            Default is 0.1.
+        additional_kwargs (dict):
+            Specify additional parameters to the request body.
+        max_retries (int):
+            How many times to retry the API call if it fails.
+            Defaults to 3.
+        timeout (float):
+            How long to wait, in seconds, for an API call before failing.
+            Defaults to 60.0.
+        reuse_client (bool):
+            Reuse the OpenAI client between requests.
+            Defaults to True.
+        default_headers (dict):
+            Override the default headers for API requests.
+            Defaults to None.
+        http_client (httpx.Client):
+            Pass in your own httpx.Client instance.
+            Defaults to None.
+        async_http_client (httpx.AsyncClient):
+            Pass in your own httpx.AsyncClient instance.
+            Defaults to None.
 
     Examples:
         `pip install llama-index-llms-openai-like`

llama-index-integrations/llms/llama-index-llms-openai-like/pyproject.toml

@@ -27,7 +27,7 @@ exclude = ["**/BUILD"]
 license = "MIT"
 name = "llama-index-llms-openai-like"
 readme = "README.md"
-version = "0.3.3"
+version = "0.3.4"
 
 [tool.poetry.dependencies]
 python = ">=3.9,<4.0"