diff --git a/utils/usage.ipynb b/utils/usage.ipynb
new file mode 100644
index 00000000..7bd733b5
--- /dev/null
+++ b/utils/usage.ipynb
@@ -0,0 +1,261 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "True"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import os\n",
+    "import sys\n",
+    "from sambanova_endpoint import SambaNovaEndpoint, SambaverseEndpoint\n",
+    "from dotenv import load_dotenv\n",
+    "import json\n",
+    "load_dotenv(\"../.env\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# SambaStudio endpoint"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Non streaming"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "llm = SambaNovaEndpoint(\n",
+    "            streaming=False,\n",
+    "            model_kwargs={\n",
+    "                \"do_sample\": True, \n",
+    "                \"temperature\": 0.01,\n",
+    "                \"max_tokens_to_generate\": 256,\n",
+    "                # \"repetition_penalty\": {\"type\": \"float\", \"value\": \"1\"},\n",
+    "                # \"top_k\": {\"type\": \"int\", \"value\": \"50\"},\n",
+    "                # \"top_logprobs\": {\"type\": \"int\", \"value\": \"0\"},\n",
+    "                # \"top_p\": {\"type\": \"float\", \"value\": \"1\"}\n",
+    "            }\n",
+    "        ) \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "\"of a man who was once a great warrior, but is now a humble farmer.\\n\\nThe sun sets on the fields, a weary farmer bends to his work. Once a great warrior, he fought for his people's freedom. Now, he fights for his family's survival. His calloused hands, a testament to his past, still hold the strength of a warrior. But his heart, once filled with anger and vengeance, now holds only love and hope.\""
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "llm.invoke(\"tell me a 50 word tale\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Streaming"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "llm=SambaNovaEndpoint(\n",
+    "    streaming=True,\n",
+    "    model_kwargs={\n",
+    "        \"do_sample\": True,\n",
+    "        \"max_tokens_to_generate\": 256,\n",
+    "        \"temperature\": 0.01,\n",
+    "        # \"repetition_penalty\": {\"type\": \"float\", \"value\": \"1\"},\n",
+    "        # \"top_k\": {\"type\": \"int\", \"value\": \"50\"},\n",
+    "        # \"top_logprobs\": {\"type\": \"int\", \"value\": \"0\"},\n",
+    "        # \"top_p\": {\"type\": \"float\", \"value\": \"1\"}\n",
+    "        }\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      " of a man who was once a great warrior, but is now a humble farmer.\n",
+      "\n",
+      "The sun sets on the fields, a weary farmer bends to his work. Once a great warrior, he fought for his people's freedom. Now, he fights for his family's survival. His calloused hands, a testament to his past, still hold the strength of a warrior. But his heart, once filled with anger and vengeance, now holds only love and hope."
+     ]
+    }
+   ],
+   "source": [
+    "for chunk in llm.stream(\"tell me a 50 word tale\"):\n",
+    "    data = json.loads(json.loads(chunk)[\"data\"])[\"stream_token\"]\n",
+    "    print(data, end=\"\", flush=True)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Sambaverse\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Non Streaming"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "llm = SambaverseEndpoint(\n",
+    "        sambaverse_model_name=\"Meta/llama-2-7b-chat-hf\",\n",
+    "        model_kwargs={\n",
+    "            \"do_sample\": True, \n",
+    "            \"max_tokens_to_generate\": 256,\n",
+    "            \"temperature\": 0.01,\n",
+    "            \"process_prompt\": True,\n",
+    "            \"select_expert\": \"llama-2-7b-chat-hf\"\n",
+    "            #\"stop_sequences\": { \"type\":\"str\", \"value\":\"\"},\n",
+    "            # \"repetition_penalty\": {\"type\": \"float\", \"value\": \"1\"},\n",
+    "            # \"top_k\": {\"type\": \"int\", \"value\": \"50\"},\n",
+    "            # \"top_p\": {\"type\": \"float\", \"value\": \"1\"}\n",
+    "        }\n",
+    "    ) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "' Sure! Here is a 50-word tale:\\n\\nThe cat purred contentedly on my lap, pawing at my hand with a gentle mew.'"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "llm.invoke(\"tell me a 50 word tale\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Streaming"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "llm = SambaverseEndpoint(\n",
+    "        streaming=True,\n",
+    "        sambaverse_model_name=\"Meta/llama-2-7b-chat-hf\",\n",
+    "        model_kwargs={\n",
+    "            \"do_sample\": True, \n",
+    "            \"max_tokens_to_generate\": 256,\n",
+    "            \"temperature\": 0.01,\n",
+    "            \"process_prompt\": True,\n",
+    "            \"select_expert\": \"llama-2-7b-chat-hf\"\n",
+    "            #\"stop_sequences\": { \"type\":\"str\", \"value\":\"\"},\n",
+    "            # \"repetition_penalty\": {\"type\": \"float\", \"value\": \"1\"},\n",
+    "            # \"top_k\": {\"type\": \"int\", \"value\": \"50\"},\n",
+    "            # \"top_p\": {\"type\": \"float\", \"value\": \"1\"}\n",
+    "        }\n",
+    "    ) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "  Sure! Here is a 50-word tale:\n",
+      "\n",
+      "The cat purred contentedly on my lap, pawing at my hand with a gentle mew."
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "for chunk in llm.stream(\"tell me a 50 word tale\"):\n",
+    "    data = json.loads(json.loads(chunk)[\"data\"])[\"stream_token\"]\n",
+    "    print(data, end=\"\", flush=True)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "peenv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}