litellm/cookbook/LiteLLM_CometAPI.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LiteLLM CometAPI Cookbook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install litellm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "# CometAPI keys can be provided as COMETAPI_KEY or COMETAPI_API_KEY.\n",
    "api_key = os.getenv(\"COMETAPI_KEY\") or os.getenv(\"COMETAPI_API_KEY\")\n",
    "if api_key is None:\n",
    "    raise RuntimeError(\"Set COMETAPI_KEY or COMETAPI_API_KEY before running this notebook.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Chat Completion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from litellm import completion\n",
    "\n",
    "response = completion(\n",
    "    model=\"cometapi/gpt-5.5\",\n",
    "    messages=[{\"role\": \"user\", \"content\": \"Write Python code that prints hi.\"}],\n",
    "    max_tokens=128,\n",
    "    api_key=api_key,\n",
    ")\n",
    "\n",
    "print(response.choices[0].message.content)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Streaming Chat Completion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "messages = [{\"role\": \"user\", \"content\": \"Count to three.\"}]\n",
    "response = completion(\n",
    "    model=\"cometapi/gpt-5.5\",\n",
    "    messages=messages,\n",
    "    max_tokens=64,\n",
    "    stream=True,\n",
    "    api_key=api_key,\n",
    ")\n",
    "\n",
    "for part in response:\n",
    "    print(part.choices[0].delta.content or \"\", end=\"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Async Chat Completion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from litellm import acompletion\n",
    "\n",
    "async def run_async_chat():\n",
    "    response = await acompletion(\n",
    "        model=\"cometapi/gpt-5.5\",\n",
    "        messages=[{\"content\": \"Hello, how are you?\", \"role\": \"user\"}],\n",
    "        max_tokens=64,\n",
    "        api_key=api_key,\n",
    "    )\n",
    "    return response\n",
    "\n",
    "response = await run_async_chat()\n",
    "print(response.choices[0].message.content)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Async Streaming Chat Completion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "async def run_async_streaming_chat():\n",
    "    response = await acompletion(\n",
    "        model=\"cometapi/gpt-5.5\",\n",
    "        messages=[{\"content\": \"Stream one short sentence.\", \"role\": \"user\"}],\n",
    "        max_tokens=64,\n",
    "        stream=True,\n",
    "        api_key=api_key,\n",
    "    )\n",
    "    async for chunk in response:\n",
    "        print(chunk.choices[0].delta.content or \"\", end=\"\")\n",
    "\n",
    "await run_async_streaming_chat()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import litellm\n",
    "\n",
    "response = litellm.embedding(\n",
    "    model=\"cometapi/text-embedding-3-small\",\n",
    "    input=[\"LiteLLM routes this embedding request through CometAPI.\"],\n",
    "    api_key=api_key,\n",
    ")\n",
    "\n",
    "print(len(response.data[0][\"embedding\"]))\n",
    "print(response.usage)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Async Embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "async def run_async_embedding():\n",
    "    response = await litellm.aembedding(\n",
    "        model=\"cometapi/text-embedding-3-small\",\n",
    "        input=\"Your text string\",\n",
    "        api_key=api_key,\n",
    "    )\n",
    "    return response\n",
    "\n",
    "response = await run_async_embedding()\n",
    "print(len(response.data[0][\"embedding\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Image Generation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response = litellm.image_generation(\n",
    "    model=\"cometapi/gpt-image-2\",\n",
    "    prompt=\"A small comet over a clean API diagram\",\n",
    "    size=\"1024x1024\",\n",
    "    api_key=api_key,\n",
    ")\n",
    "\n",
    "image = response.data[0]\n",
    "print(image.url or image.b64_json[:64])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Async Image Generation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "async def run_async_image_generation():\n",
    "    response = await litellm.aimage_generation(\n",
    "        model=\"cometapi/gpt-image-2\",\n",
    "        prompt=\"A cute baby sea otter\",\n",
    "        size=\"1024x1024\",\n",
    "        api_key=api_key,\n",
    "    )\n",
    "    return response\n",
    "\n",
    "response = await run_async_image_generation()\n",
    "image = response.data[0]\n",
    "print(image.url or image.b64_json[:64])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Audio Speech"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import io\n",
    "\n",
    "speech_response = litellm.speech(\n",
    "    model=\"cometapi/tts-1\",\n",
    "    input=\"LiteLLM can route speech generation through CometAPI.\",\n",
    "    voice=\"alloy\",\n",
    "    api_key=api_key,\n",
    ")\n",
    "\n",
    "speech_audio = io.BytesIO(speech_response.content)\n",
    "speech_audio.name = \"speech.mp3\"\n",
    "\n",
    "print(len(speech_response.content))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Audio Transcription"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "speech_audio.seek(0)\n",
    "transcription_response = litellm.transcription(\n",
    "    model=\"cometapi/whisper-1\",\n",
    "    file=speech_audio,\n",
    "    api_key=api_key,\n",
    ")\n",
    "\n",
    "print(transcription_response.text)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Moderations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "moderation_response = litellm.moderation(\n",
    "    model=\"omni-moderation-latest\",\n",
    "    custom_llm_provider=\"cometapi\",\n",
    "    input=\"I want to build a safe application.\",\n",
    "    api_key=api_key,\n",
    ")\n",
    "\n",
    "print(moderation_response.results[0].categories)\n",
    "print(moderation_response.results[0].category_scores)\n"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.12.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}