From e593e017d2a2eb62fbf94c89aa35660e743c353f Mon Sep 17 00:00:00 2001
From: Yuki Watanabe <31463517+B-Step62@users.noreply.github.com>
Date: Thu, 31 Oct 2024 00:56:55 +0900
Subject: [PATCH] Update compatibility table for ChatDatabricks (#27676)

`ChatDatabricks` added support for structured output and JSON mode in
the last release. This PR updates the feature table accordingly.

Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
---
 docs/docs/integrations/chat/databricks.ipynb | 53 +++++++++++++++++++-
 1 file changed, 51 insertions(+), 2 deletions(-)

diff --git a/docs/docs/integrations/chat/databricks.ipynb b/docs/docs/integrations/chat/databricks.ipynb
index 071a9f16881..5b66d2b91ff 100644
--- a/docs/docs/integrations/chat/databricks.ipynb
+++ b/docs/docs/integrations/chat/databricks.ipynb
@@ -36,7 +36,7 @@
     "### Model features\n",
     "| [Tool calling](/docs/how_to/tool_calling/) | [Structured output](/docs/how_to/structured_output/) | JSON mode | [Image input](/docs/how_to/multimodal_inputs/) | Audio input | Video input | [Token-level streaming](/docs/how_to/chat_streaming/) | Native async | [Token usage](/docs/how_to/chat_token_usage_tracking/) | [Logprobs](/docs/how_to/logprobs/) |\n",
     "| :---: | :---: | :---: | :---: |  :---: | :---: | :---: | :---: | :---: | :---: |\n",
-    "| ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |  ✅ | ✅ | ✅ | ❌ | \n",
+    "| ✅ | ✅ | ✅ | ❌ | ❌ | ❌ |  ✅ | ✅ | ✅ | ❌ | \n",
     "\n",
     "### Supported Methods\n",
     "\n",
@@ -289,6 +289,55 @@
     "await asyncio.gather(*futures)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Tool calling\n",
+    "\n",
+    "ChatDatabricks supports OpenAI-compatible tool calling API that lets you describe tools and their arguments, and have the model return a JSON object with a tool to invoke and the inputs to that tool. tool-calling is extremely useful for building tool-using chains and agents, and for getting structured outputs from models more generally.\n",
+    "\n",
+    "With `ChatDatabricks.bind_tools`, we can easily pass in Pydantic classes, dict schemas, LangChain tools, or even functions as tools to the model. Under the hood these are converted to the OpenAI-compatible tool schemas, which looks like:\n",
+    "\n",
+    "```\n",
+    "{\n",
+    "    \"name\": \"...\",\n",
+    "    \"description\": \"...\",\n",
+    "    \"parameters\": {...}  # JSONSchema\n",
+    "}\n",
+    "```\n",
+    "\n",
+    "and passed in every model invocation."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pydantic import BaseModel, Field\n",
+    "\n",
+    "\n",
+    "class GetWeather(BaseModel):\n",
+    "    \"\"\"Get the current weather in a given location\"\"\"\n",
+    "\n",
+    "    location: str = Field(..., description=\"The city and state, e.g. San Francisco, CA\")\n",
+    "\n",
+    "\n",
+    "class GetPopulation(BaseModel):\n",
+    "    \"\"\"Get the current population in a given location\"\"\"\n",
+    "\n",
+    "    location: str = Field(..., description=\"The city and state, e.g. San Francisco, CA\")\n",
+    "\n",
+    "\n",
+    "llm_with_tools = chat_model.bind_tools([GetWeather, GetPopulation])\n",
+    "ai_msg = llm_with_tools.invoke(\n",
+    "    \"Which city is hotter today and which is bigger: LA or NY?\"\n",
+    ")\n",
+    "print(ai_msg.tool_calls)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -466,7 +515,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "Python 3",
    "language": "python",
    "name": "python3"
   },