community: SamabaStudio Tool Calling and Structured Output (#28025)

Description: Add tool calling and structured output support for
SambaStudio chat models, docs included

---------

Co-authored-by: Erick Friis <erick@langchain.dev>

docs/docs/integrations/chat/sambastudio.ipynb

@@ -34,7 +34,7 @@
     "\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",
     "## Setup\n",
     "\n",
@@ -119,20 +119,20 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
     "from langchain_community.chat_models.sambanova import ChatSambaStudio\n",
     "\n",
     "llm = ChatSambaStudio(\n",
-    "    model=\"Meta-Llama-3-70B-Instruct-4096\",  # set if using a CoE endpoint\n",
+    "    model=\"Meta-Llama-3-70B-Instruct-4096\",  # set if using a Bundle endpoint\n",
     "    max_tokens=1024,\n",
     "    temperature=0.7,\n",
     "    top_k=1,\n",
     "    top_p=0.01,\n",
     "    do_sample=True,\n",
-    "    process_prompt=\"True\",  # set if using a CoE endpoint\n",
+    "    process_prompt=\"True\",  # set if using a Bundle endpoint\n",
     ")"
    ]
   },
@@ -349,6 +349,134 @@
     "    print(chunk.content, end=\"\", flush=True)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Tool calling"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from datetime import datetime\n",
+    "\n",
+    "from langchain_core.messages import HumanMessage, SystemMessage, ToolMessage\n",
+    "from langchain_core.tools import tool\n",
+    "\n",
+    "\n",
+    "@tool\n",
+    "def get_time(kind: str = \"both\") -> str:\n",
+    "    \"\"\"Returns current date, current time or both.\n",
+    "    Args:\n",
+    "        kind: date, time or both\n",
+    "    \"\"\"\n",
+    "    if kind == \"date\":\n",
+    "        date = datetime.now().strftime(\"%m/%d/%Y\")\n",
+    "        return f\"Current date: {date}\"\n",
+    "    elif kind == \"time\":\n",
+    "        time = datetime.now().strftime(\"%H:%M:%S\")\n",
+    "        return f\"Current time: {time}\"\n",
+    "    else:\n",
+    "        date = datetime.now().strftime(\"%m/%d/%Y\")\n",
+    "        time = datetime.now().strftime(\"%H:%M:%S\")\n",
+    "        return f\"Current date: {date}, Current time: {time}\"\n",
+    "\n",
+    "\n",
+    "tools = [get_time]\n",
+    "\n",
+    "\n",
+    "def invoke_tools(tool_calls, messages):\n",
+    "    available_functions = {tool.name: tool for tool in tools}\n",
+    "    for tool_call in tool_calls:\n",
+    "        selected_tool = available_functions[tool_call[\"name\"]]\n",
+    "        tool_output = selected_tool.invoke(tool_call[\"args\"])\n",
+    "        print(f\"Tool output: {tool_output}\")\n",
+    "        messages.append(ToolMessage(tool_output, tool_call_id=tool_call[\"id\"]))\n",
+    "    return messages"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "llm_with_tools = llm.bind_tools(tools=tools)\n",
+    "messages = [\n",
+    "    HumanMessage(\n",
+    "        content=\"I need to schedule a meeting for two weeks from today. Can you tell me the exact date of the meeting?\"\n",
+    "    )\n",
+    "]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Intermediate model response: [{'name': 'get_time', 'args': {'kind': 'date'}, 'id': 'call_4092d5dd21cd4eb494', 'type': 'tool_call'}]\n",
+      "Tool output: Current date: 11/07/2024\n",
+      "final response: The meeting will be exactly two weeks from today, which would be 25/07/2024.\n"
+     ]
+    }
+   ],
+   "source": [
+    "response = llm_with_tools.invoke(messages)\n",
+    "while len(response.tool_calls) > 0:\n",
+    "    print(f\"Intermediate model response: {response.tool_calls}\")\n",
+    "    messages.append(response)\n",
+    "    messages = invoke_tools(response.tool_calls, messages)\n",
+    "response = llm_with_tools.invoke(messages)\n",
+    "\n",
+    "print(f\"final response: {response.content}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Structured Outputs"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Joke(setup='Why did the cat join a band?', punchline='Because it wanted to be the purr-cussionist!')"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from pydantic import BaseModel, Field\n",
+    "\n",
+    "\n",
+    "class Joke(BaseModel):\n",
+    "    \"\"\"Joke to tell user.\"\"\"\n",
+    "\n",
+    "    setup: str = Field(description=\"The setup of the joke\")\n",
+    "    punchline: str = Field(description=\"The punchline to the joke\")\n",
+    "\n",
+    "\n",
+    "structured_llm = llm.with_structured_output(Joke)\n",
+    "\n",
+    "structured_llm.invoke(\"Tell me a joke about cats\")"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},