Skip to content

Instantly share code, notes, and snippets.

@ninehills

ninehills/chatglm-openai-api.ipynb

Last active April 16, 2024 01:15
Show Gist options
  • Save ninehills/f675e05a595db39b491fab1d661fced6 to your computer and use it in GitHub Desktop.
Save ninehills/f675e05a595db39b491fab1d661fced6 to your computer and use it in GitHub Desktop.
Download ZIP
chatglm-openai-api.ipynb
Display the source blob
Display the rendered blob
Raw
chatglm-openai-api.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": []
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
},
"accelerator": "GPU",
"gpuClass": "standard"
},
"cells": [
{
"cell_type": "markdown",
"source": [
"# **本地版本的在： https://github.com/ninehills/chatglm-openai-api 后续主要维护本地版本！！！**"
],
"metadata": {
"id": "wwkHnLlNRAgm"
}
},
{
"cell_type": "markdown",
"source": [
"## 安装依赖"
],
"metadata": {
"id": "MZyEDzEHyDcN"
}
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "eNbQgYuLxp-3",
"outputId": "61a6f8a0-5c12-46c4-d4b8-0d71c7d8858d"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
"Requirement already satisfied: protobuf==3.20.0 in /usr/local/lib/python3.9/dist-packages (3.20.0)\n",
"Requirement already satisfied: transformers==4.27.1 in /usr/local/lib/python3.9/dist-packages (4.27.1)\n",
"Requirement already satisfied: icetk in /usr/local/lib/python3.9/dist-packages (0.0.4)\n",
"Requirement already satisfied: cpm_kernels in /usr/local/lib/python3.9/dist-packages (1.0.11)\n",
"Requirement already satisfied: torch in /usr/local/lib/python3.9/dist-packages (2.0.0+cu118)\n",
"Requirement already satisfied: filelock in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (3.11.0)\n",
"Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (1.22.4)\n",
"Requirement already satisfied: requests in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (2.27.1)\n",
"Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (4.65.0)\n",
"Requirement already satisfied: tokenizers!=0.11.3,<0.14,>=0.11.1 in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (0.13.3)\n",
"Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (6.0)\n",
"Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (2022.10.31)\n",
"Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (23.1)\n",
"Requirement already satisfied: huggingface-hub<1.0,>=0.11.0 in /usr/local/lib/python3.9/dist-packages (from transformers==4.27.1) (0.13.4)\n",
"Requirement already satisfied: sentencepiece in /usr/local/lib/python3.9/dist-packages (from icetk) (0.1.98)\n",
"Requirement already satisfied: torchvision in /usr/local/lib/python3.9/dist-packages (from icetk) (0.15.1+cu118)\n",
"Requirement already satisfied: networkx in /usr/local/lib/python3.9/dist-packages (from torch) (3.1)\n",
"Requirement already satisfied: jinja2 in /usr/local/lib/python3.9/dist-packages (from torch) (3.1.2)\n",
"Requirement already satisfied: triton==2.0.0 in /usr/local/lib/python3.9/dist-packages (from torch) (2.0.0)\n",
"Requirement already satisfied: typing-extensions in /usr/local/lib/python3.9/dist-packages (from torch) (4.5.0)\n",
"Requirement already satisfied: sympy in /usr/local/lib/python3.9/dist-packages (from torch) (1.11.1)\n",
"Requirement already satisfied: cmake in /usr/local/lib/python3.9/dist-packages (from triton==2.0.0->torch) (3.25.2)\n",
"Requirement already satisfied: lit in /usr/local/lib/python3.9/dist-packages (from triton==2.0.0->torch) (16.0.1)\n",
"Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.9/dist-packages (from jinja2->torch) (2.1.2)\n",
"Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.9/dist-packages (from requests->transformers==4.27.1) (2.0.12)\n",
"Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.9/dist-packages (from requests->transformers==4.27.1) (2022.12.7)\n",
"Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.9/dist-packages (from requests->transformers==4.27.1) (3.4)\n",
"Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.9/dist-packages (from requests->transformers==4.27.1) (1.26.15)\n",
"Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.9/dist-packages (from sympy->torch) (1.3.0)\n",
"Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.9/dist-packages (from torchvision->icetk) (8.4.0)\n",
"Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
"Requirement already satisfied: fastapi in /usr/local/lib/python3.9/dist-packages (0.95.1)\n",
"Requirement already satisfied: pydantic in /usr/local/lib/python3.9/dist-packages (1.10.7)\n",
"Requirement already satisfied: uvicorn in /usr/local/lib/python3.9/dist-packages (0.21.1)\n",
"Requirement already satisfied: sse_starlette in /usr/local/lib/python3.9/dist-packages (1.3.4)\n",
"Requirement already satisfied: pyngrok in /usr/local/lib/python3.9/dist-packages (6.0.0)\n",
"Requirement already satisfied: nest-asyncio in /usr/local/lib/python3.9/dist-packages (1.5.6)\n",
"Requirement already satisfied: starlette<0.27.0,>=0.26.1 in /usr/local/lib/python3.9/dist-packages (from fastapi) (0.26.1)\n",
"Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.9/dist-packages (from pydantic) (4.5.0)\n",
"Requirement already satisfied: h11>=0.8 in /usr/local/lib/python3.9/dist-packages (from uvicorn) (0.14.0)\n",
"Requirement already satisfied: click>=7.0 in /usr/local/lib/python3.9/dist-packages (from uvicorn) (8.1.3)\n",
"Requirement already satisfied: PyYAML in /usr/local/lib/python3.9/dist-packages (from pyngrok) (6.0)\n",
"Requirement already satisfied: anyio<5,>=3.4.0 in /usr/local/lib/python3.9/dist-packages (from starlette<0.27.0,>=0.26.1->fastapi) (3.6.2)\n",
"Requirement already satisfied: sniffio>=1.1 in /usr/local/lib/python3.9/dist-packages (from anyio<5,>=3.4.0->starlette<0.27.0,>=0.26.1->fastapi) (1.3.0)\n",
"Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.9/dist-packages (from anyio<5,>=3.4.0->starlette<0.27.0,>=0.26.1->fastapi) (3.4)\n"
]
}
],
"source": [
"import locale\n",
"locale.getpreferredencoding = lambda: \"UTF-8\"\n",
"\n",
"!pip install protobuf==3.20.0 transformers==4.27.1 icetk cpm_kernels torch\n",
"!pip install fastapi pydantic uvicorn sse_starlette pyngrok nest-asyncio"
]
},
{
"cell_type": "markdown",
"source": [
"## 环境配置"
],
"metadata": {
"id": "VaY7_gNOzdoG"
}
},
{
"cell_type": "code",
"source": [
"import torch\n",
"if torch.cuda.is_available() == False:\n",
" print(\"请在右上方 Colab 运行时类型中，选择 GPU 类型的运行时。\")"
],
"metadata": {
"id": "rVBP-rhiQhT5"
},
"execution_count": 2,
"outputs": []
},
{
"cell_type": "code",
"source": [
"chatglm_models = [\n",
" \"THUDM/chatglm-6b\", # 原始模型\n",
" \"THUDM/chatglm-6b-int8\", # int8 量化\n",
" \"THUDM/chatglm-6b-int4\", # int4 量化\n",
"]\n",
"\n",
"CHATGLM_MODEL = \"THUDM/chatglm-6b-int4\"\n",
"\n",
"# GPU/CPU\n",
"RUNNING_DEVICE = \"GPU\"\n",
"\n",
"# API_TOKEN\n",
"TOKEN = \"token1\"\n"
],
"metadata": {
"id": "JJOlpnVOyIeG"
},
"execution_count": 3,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## 启动模型"
],
"metadata": {
"id": "CFzHRCMYz08N"
}
},
{
"cell_type": "code",
"source": [
"from transformers import AutoModel, AutoTokenizer\n",
"\n",
"def init_chatglm(model_name: str, running_device: str):\n",
" tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)\n",
" model = AutoModel.from_pretrained(model_name, trust_remote_code=True)\n",
"\n",
" if running_device == \"GPU\":\n",
" model = model.half().cuda()\n",
" else:\n",
" model = model.float()\n",
" model.eval()\n",
" return tokenizer, model\n",
"\n",
"tokenizer, model = init_chatglm(CHATGLM_MODEL, RUNNING_DEVICE)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "VIT1nTv_1a4X",
"outputId": "fe2fe031-4f83-4c20-b882-19682c528647"
},
"execution_count": 4,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
"Explicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.\n",
"Explicitly passing a `revision` is encouraged when loading a configuration with custom code to ensure no malicious code has been contributed in a newer revision.\n",
"Explicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.\n"
]
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"No compiled kernel found.\n",
"Compiling kernels : /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4/e02ba894cf18f3fd9b2526c795f983683c4ec732/quantization_kernels.c\n",
"Compiling gcc -O3 -fPIC -std=c99 /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4/e02ba894cf18f3fd9b2526c795f983683c4ec732/quantization_kernels.c -shared -o /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4/e02ba894cf18f3fd9b2526c795f983683c4ec732/quantization_kernels.so\n",
"Load kernel : /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4/e02ba894cf18f3fd9b2526c795f983683c4ec732/quantization_kernels.so\n",
"Using quantization cache\n",
"Applying quantization to glm layers\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"## 测试模型"
],
"metadata": {
"id": "b5RuRcsD3hPw"
}
},
{
"cell_type": "code",
"source": [
"response, history = model.chat(tokenizer, \"你好\", history=[])\n",
"print(response)\n",
"print(history)\n",
"response, history = model.chat(tokenizer, \"很高兴认识你\", history=history)\n",
"print(response)\n",
"print(history)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "ZFY3ju-N3gk1",
"outputId": "184af497-1c9a-46bc-b76c-c4847bebf1ee"
},
"execution_count": 5,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
"WARNING:transformers_modules.THUDM.chatglm-6b-int4.e02ba894cf18f3fd9b2526c795f983683c4ec732.modeling_chatglm:The dtype of attention mask (torch.int64) is not bool\n"
]
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"你好👋！我是人工智能助手 ChatGLM-6B，很高兴见到你，欢迎问我任何问题。\n",
"[('你好', '你好👋！我是人工智能助手 ChatGLM-6B，很高兴见到你，欢迎问我任何问题。')]\n",
"非常感谢你的欢迎！作为一名人工智能助手，我的任务是帮助人们和计算机之间进行更好的交流和理解。我很高兴能够提供帮助和回答问题，如果有任何疑问，随时都可以问我。\n",
"[('你好', '你好👋！我是人工智能助手 ChatGLM-6B，很高兴见到你，欢迎问我任何问题。'), ('很高兴认识你', '非常感谢你的欢迎！作为一名人工智能助手，我的任务是帮助人们和计算机之间进行更好的交流和理解。我很高兴能够提供帮助和回答问题，如果有任何疑问，随时都可以问我。')]\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"## 启动服务器（with tunnel）"
],
"metadata": {
"id": "LnexGhqK39NJ"
}
},
{
"cell_type": "code",
"source": [
"!wget https://github.com/ninehills/chatgpt-web/releases/download/1.0/dist.tgz -O dist.tgz\n",
"!tar zxvf dist.tgz\n",
"!mv ./dist/index.html ./dist/assets/"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "Up0KMv4vRtsP",
"outputId": "6f3c9eaf-e1b2-4666-89db-c4664c872563"
},
"execution_count": 8,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"--2023-04-21 11:14:14-- https://github.com/ninehills/chatgpt-web/releases/download/1.0/dist.tgz\n",
"Resolving github.com (github.com)... 192.30.255.112\n",
"Connecting to github.com (github.com)|192.30.255.112|:443... connected.\n",
"HTTP request sent, awaiting response... 302 Found\n",
"Location: https://objects.githubusercontent.com/github-production-release-asset-2e65be/630857471/b1a34dc5-00fe-4969-b2dc-47af921d7dbf?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230421%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230421T111414Z&X-Amz-Expires=300&X-Amz-Signature=07ddaa9df7f8be4f3c4c4f484d2bdee8034f2963dcaaa2ea567ade506647179f&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=630857471&response-content-disposition=attachment%3B%20filename%3Ddist.tgz&response-content-type=application%2Foctet-stream [following]\n",
"--2023-04-21 11:14:14-- https://objects.githubusercontent.com/github-production-release-asset-2e65be/630857471/b1a34dc5-00fe-4969-b2dc-47af921d7dbf?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230421%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230421T111414Z&X-Amz-Expires=300&X-Amz-Signature=07ddaa9df7f8be4f3c4c4f484d2bdee8034f2963dcaaa2ea567ade506647179f&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=630857471&response-content-disposition=attachment%3B%20filename%3Ddist.tgz&response-content-type=application%2Foctet-stream\n",
"Resolving objects.githubusercontent.com (objects.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.108.133, ...\n",
"Connecting to objects.githubusercontent.com (objects.githubusercontent.com)|185.199.109.133|:443... connected.\n",
"HTTP request sent, awaiting response... 200 OK\n",
"Length: 104402 (102K) [application/octet-stream]\n",
"Saving to: ‘dist.tgz’\n",
"\n",
"dist.tgz 100%[===================>] 101.96K --.-KB/s in 0.02s \n",
"\n",
"2023-04-21 11:14:14 (6.55 MB/s) - ‘dist.tgz’ saved [104402/104402]\n",
"\n",
"./dist/\n",
"./dist/index.html\n",
"./dist/assets/\n",
"./dist/assets/index-50bc3425.css\n",
"./dist/assets/index-8effcbc5.js\n",
"./dist/assets/logo-66789ed1.svg\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"import torch\n",
"from fastapi import FastAPI, Request, status, HTTPException\n",
"from fastapi.responses import JSONResponse, HTMLResponse\n",
"from pydantic import BaseModel\n",
"from sse_starlette.sse import EventSourceResponse\n",
"from fastapi.responses import RedirectResponse\n",
"from fastapi.staticfiles import StaticFiles\n",
"from fastapi.middleware.cors import CORSMiddleware\n",
"import uvicorn\n",
"import json\n",
"from typing import List, Optional\n",
"\n",
"\n",
"# 参考 https://github.com/josStorer/selfhostedAI/blob/master/main.py\n",
"\n",
"def torch_gc():\n",
" if torch.cuda.is_available():\n",
" with torch.cuda.device(0):\n",
" torch.cuda.empty_cache()\n",
" torch.cuda.ipc_collect()\n",
"\n",
"\n",
"app = FastAPI()\n",
"\n",
"app.add_middleware(\n",
" CORSMiddleware,\n",
" allow_origins=['*'],\n",
" allow_credentials=True,\n",
" allow_methods=['*'],\n",
" allow_headers=['*'],\n",
")\n",
"\n",
"app.mount(\"/assets\", StaticFiles(directory=\"dist/assets\"), name=\"assets\")\n",
"\n",
"\n",
"class Message(BaseModel):\n",
" role: str\n",
" content: str\n",
"\n",
"\n",
"class Body(BaseModel):\n",
" messages: List[Message]\n",
" model: str\n",
" stream: Optional[bool] = False\n",
" max_tokens: Optional[int] = 256\n",
" temperature: Optional[float] = 0.95\n",
" top_p: Optional[float] = 0.7\n",
"\n",
"\n",
"\n",
"@app.get(\"/\")\n",
"def read_root():\n",
" return RedirectResponse(\"/assets/index.html\")\n",
"\n",
"\n",
"@app.get(\"/v1/models\")\n",
"def get_models():\n",
" return {\"data\": [\n",
" {\n",
" \"created\": 1677610602,\n",
" \"id\": \"gpt-3.5-turbo\",\n",
" \"object\": \"model\",\n",
" \"owned_by\": \"openai\",\n",
" \"permission\": [\n",
" {\n",
" \"created\": 1680818747,\n",
" \"id\": \"modelperm-fTUZTbzFp7uLLTeMSo9ks6oT\",\n",
" \"object\": \"model_permission\",\n",
" \"allow_create_engine\": False,\n",
" \"allow_sampling\": True,\n",
" \"allow_logprobs\": True,\n",
" \"allow_search_indices\": False,\n",
" \"allow_view\": True,\n",
" \"allow_fine_tuning\": False,\n",
" \"organization\": \"*\",\n",
" \"group\": None,\n",
" \"is_blocking\": False\n",
" }\n",
" ],\n",
" \"root\": \"gpt-3.5-turbo\",\n",
" \"parent\": None,\n",
" },\n",
" ],\n",
" \"object\": \"list\"\n",
" }\n",
"\n",
"def generate_response(content: str):\n",
" return {\n",
" \"id\": \"chatcmpl-77PZm95TtxE0oYLRx3cxa6HtIDI7s\",\n",
" \"object\": \"chat.completion\",\n",
" \"created\": 1682000966,\n",
" \"model\": \"gpt-3.5-turbo-0301\",\n",
" \"usage\": {\n",
" \"prompt_tokens\": 10,\n",
" \"completion_tokens\": 10,\n",
" \"total_tokens\": 20,\n",
" },\n",
" \"choices\": [{\n",
" \"message\": {\"role\": \"assistant\", \"content\": content}, \"finish_reason\": \"stop\", \"index\": 0}\n",
" ]\n",
" }\n",
"\n",
"def generate_stream_response_start():\n",
" return {\"id\":\"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\",\"object\":\"chat.completion.chunk\",\"created\":1682004627,\"model\":\"gpt-3.5-turbo-0301\",\"choices\":[{\"delta\":{\"role\":\"assistant\"},\"index\":0,\"finish_reason\":None}]}\n",
"\n",
"def generate_stream_response(content: str):\n",
" return {\n",
" \"id\":\"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\",\n",
" \"object\":\"chat.completion.chunk\",\n",
" \"created\":1682004627,\n",
" \"model\":\"gpt-3.5-turbo-0301\",\n",
" \"choices\":[{\"delta\":{\"content\":content},\"index\":0,\"finish_reason\":None}\n",
" ]}\n",
"\n",
"def generate_stream_response_stop():\n",
" return {\"id\":\"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\",\"object\":\"chat.completion.chunk\",\"created\":1682004627,\"model\":\"gpt-3.5-turbo-0301\",\"choices\":[{\"delta\":{},\"index\":0,\"finish_reason\":\"stop\"}]}\n",
"\n",
"@app.post(\"/v1/chat/completions\")\n",
"async def completions(body: Body, request: Request):\n",
" # Cancel token\n",
" #if request.headers.get(\"Authorization\").split(\" \")[1] != TOKEN:\n",
" # raise HTTPException(status.HTTP_401_UNAUTHORIZED, \"Token is wrong!\")\n",
" \n",
" torch_gc()\n",
"\n",
" question = body.messages[-1]\n",
" if question.role == 'user':\n",
" question = question.content\n",
" else:\n",
" raise HTTPException(status.HTTP_400_BAD_REQUEST, \"No Question Found\")\n",
"\n",
" history = []\n",
" user_question = ''\n",
" for message in body.messages:\n",
" if message.role == 'system':\n",
" history.append((message.content, \"OK\"))\n",
" if message.role == 'user':\n",
" user_question = message.content\n",
" elif message.role == 'assistant':\n",
" assistant_answer = message.content\n",
" history.append((user_question, assistant_answer))\n",
"\n",
" print(f\"question = {question}, history = {history}\")\n",
"\n",
" \n",
" if body.stream:\n",
" async def eval_chatglm():\n",
" sends = 0\n",
" first = True\n",
" for response, _ in model.stream_chat(\n",
" tokenizer, question, history,\n",
" temperature=body.temperature,\n",
" top_p=body.top_p,\n",
" max_length=max(2048, body.max_tokens)):\n",
" if await request.is_disconnected():\n",
" return\n",
" ret = response[sends:]\n",
" sends = len(response)\n",
" if first:\n",
" first = False\n",
" yield json.dumps(generate_stream_response_start(), ensure_ascii=False)\n",
" yield json.dumps(generate_stream_response(ret), ensure_ascii=False)\n",
" yield json.dumps(generate_stream_response_stop(), ensure_ascii=False)\n",
" yield \"[DONE]\"\n",
" return EventSourceResponse(eval_chatglm(), ping=10000)\n",
" else:\n",
" response, _ = model.chat(\n",
" tokenizer, question, history,\n",
" temperature=body.temperature,\n",
" top_p=body.top_p,\n",
" max_length=max(2048, body.max_tokens))\n",
" print(f\"response: {response}\")\n",
" return JSONResponse(content=generate_response(response))"
],
"metadata": {
"id": "tX5oiLQJ4BTX"
},
"execution_count": 11,
"outputs": []
},
{
"cell_type": "code",
"source": [
"!wget https://github.com/cloudflare/cloudflared/releases/latest/download/cloudflared-linux-amd64 -O cloudflared\n",
"!chmod a+x cloudflared\n"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "Cej2JQTKXZwR",
"outputId": "ff5ee79d-af42-4a32-ebd1-dc0d26d6815c"
},
"execution_count": 12,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"--2023-04-21 11:14:53-- https://github.com/cloudflare/cloudflared/releases/latest/download/cloudflared-linux-amd64\n",
"Resolving github.com (github.com)... 192.30.255.113\n",
"Connecting to github.com (github.com)|192.30.255.113|:443... connected.\n",
"HTTP request sent, awaiting response... 302 Found\n",
"Location: https://github.com/cloudflare/cloudflared/releases/download/2023.4.1/cloudflared-linux-amd64 [following]\n",
"--2023-04-21 11:14:53-- https://github.com/cloudflare/cloudflared/releases/download/2023.4.1/cloudflared-linux-amd64\n",
"Reusing existing connection to github.com:443.\n",
"HTTP request sent, awaiting response... 302 Found\n",
"Location: https://objects.githubusercontent.com/github-production-release-asset-2e65be/106867604/fdb83b81-e6e9-461a-bea7-838deb4b1387?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230421%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230421T111453Z&X-Amz-Expires=300&X-Amz-Signature=957bcb143be393c13ac68884edb62ed105de2954d10c8ceee08b96a4f625a9fe&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=106867604&response-content-disposition=attachment%3B%20filename%3Dcloudflared-linux-amd64&response-content-type=application%2Foctet-stream [following]\n",
"--2023-04-21 11:14:53-- https://objects.githubusercontent.com/github-production-release-asset-2e65be/106867604/fdb83b81-e6e9-461a-bea7-838deb4b1387?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230421%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230421T111453Z&X-Amz-Expires=300&X-Amz-Signature=957bcb143be393c13ac68884edb62ed105de2954d10c8ceee08b96a4f625a9fe&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=106867604&response-content-disposition=attachment%3B%20filename%3Dcloudflared-linux-amd64&response-content-type=application%2Foctet-stream\n",
"Resolving objects.githubusercontent.com (objects.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
"Connecting to objects.githubusercontent.com (objects.githubusercontent.com)|185.199.108.133|:443... connected.\n",
"HTTP request sent, awaiting response... 200 OK\n",
"Length: 36184431 (35M) [application/octet-stream]\n",
"Saving to: ‘cloudflared’\n",
"\n",
"cloudflared 100%[===================>] 34.51M 230MB/s in 0.2s \n",
"\n",
"2023-04-21 11:14:53 (230 MB/s) - ‘cloudflared’ saved [36184431/36184431]\n",
"\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"# 在 Notebook 中运行所需\n",
"import nest_asyncio\n",
"nest_asyncio.apply()\n",
"\n",
"#from pyngrok import ngrok, conf\n",
"\n",
"#ngrok.set_auth_token(os.environ[\"ngrok_token\"])\n",
"#http_tunnel = ngrok.connect(8000)\n",
"#print(http_tunnel.public_url)\n",
"\n",
"import subprocess\n",
"print(\"start cloudflared runnel\")\n",
"f = open(\"stdout\", \"w\")\n",
"p = subprocess.Popen(['./cloudflared', '--url', 'http://localhost:8000'], bufsize=0, stdout=f, stderr=subprocess.STDOUT)\n",
"\n",
"import time\n",
"\n",
"time.sleep(3)\n",
"\n",
"!grep -F trycloudflare stdout\n",
"\n",
"print(\"start app\")\n",
"import uvicorn\n",
"uvicorn.run(app, port=8000)\n",
"\n"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "6bPIXXdn8dG0",
"outputId": "7bdcd46b-631d-4f2f-b889-19d125468dbc"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"start cloudflared runnel\n",
"2023-04-21T11:14:57Z INF Requesting new quick Tunnel on trycloudflare.com...\n",
"2023-04-21T11:14:58Z INF | https://settle-debian-brand-relates.trycloudflare.com |\n"
]
},
{
"output_type": "stream",
"name": "stderr",
"text": [
"INFO: Started server process [19734]\n",
"INFO: Waiting for application startup.\n",
"INFO: Application startup complete.\n",
"INFO: Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)\n"
]
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"start app\n",
"INFO: 103.181.1.35:0 - \"GET / HTTP/1.1\" 307 Temporary Redirect\n",
"INFO: 103.181.1.35:0 - \"GET /index.html HTTP/1.1\" 200 OK\n",
"INFO: 103.181.1.35:0 - \"GET /index-50bc3425.css HTTP/1.1\" 200 OK\n",
"INFO: 103.181.1.35:0 - \"GET /index-8effcbc5.js HTTP/1.1\" 200 OK\n",
"INFO: 34.83.203.92:0 - \"GET /index.html HTTP/1.1\" 200 OK\n",
"INFO: 34.83.203.92:0 - \"GET /index.html HTTP/1.1\" 200 OK\n",
"INFO: 103.227.66.197:0 - \"GET /index.html HTTP/1.1\" 200 OK\n",
"INFO: 34.83.203.92:0 - \"GET /index.html HTTP/1.1\" 200 OK\n",
"INFO: 34.83.203.92:0 - \"GET /index-50bc3425.css HTTP/1.1\" 200 OK\n",
"INFO: 34.83.203.92:0 - \"GET /index-8effcbc5.js HTTP/1.1\" 200 OK\n",
"INFO: 103.181.1.35:0 - \"GET /logo-66789ed1.svg HTTP/1.1\" 200 OK\n",
"INFO: 103.227.66.197:0 - \"GET /index-50bc3425.css HTTP/1.1\" 200 OK\n",
"INFO: 103.227.66.197:0 - \"GET /index-8effcbc5.js HTTP/1.1\" 200 OK\n",
"INFO: 34.83.203.92:0 - \"GET /index.html/assets/logo-66789ed1.svg HTTP/1.1\" 404 Not Found\n",
"INFO: 103.227.66.197:0 - \"GET /logo-66789ed1.svg HTTP/1.1\" 200 OK\n",
"INFO: 34.83.203.92:0 - \"GET /favicon.ico HTTP/1.1\" 404 Not Found\n",
"question = 你好, history = []\n",
"response: 你好👋！我是人工智能助手 ChatGLM-6B，很高兴见到你，欢迎问我任何问题。\n",
"INFO: 103.181.1.35:0 - \"POST /v1/chat/completions HTTP/1.1\" 200 OK\n",
"question = 给我写一首夏天的诗, history = [('你好', '你好👋！我是人工智能助手 ChatGLM-6B，很高兴见到你，欢迎问我任何问题。')]\n",
"response: 当太阳升起，\n",
"夏日的清新空气，\n",
"在微风中拂过，\n",
"树影摇曳，鸟儿欢唱。\n",
"\n",
"蝉鸣声不断，\n",
"在树上跳跃，\n",
"游泳池的水清澈，\n",
"闪闪发光。\n",
"\n",
"在海滩上，\n",
"沙子温暖，\n",
"阳光照耀，\n",
"海浪拍打着岸边，\n",
"人们欢笑着，\n",
"享受这美好的夏日时光。\n",
"\n",
"这是一首关于夏天的诗，\n",
"一段关于享受和欢乐的记忆，\n",
"在夏天的阳光下，\n",
"我们享受着生命的美好。\n",
"INFO: 103.181.1.35:0 - \"POST /v1/chat/completions HTTP/1.1\" 200 OK\n",
"INFO: 103.181.1.35:0 - \"GET /v1/models HTTP/1.1\" 200 OK\n",
"INFO: 103.181.1.35:0 - \"GET /v1/models HTTP/1.1\" 200 OK\n",
"INFO: 103.181.1.35:0 - \"GET /v1/models HTTP/1.1\" 200 OK\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"## 访问\n",
"### 页面：\n",
"\n",
"请在上一个代码块的输出中找到类似于 https://submitted-skilled-impose-wheat.trycloudflare.com 的地址，在浏览器打开即可。\n",
"\n",
"注意可能需要登上10s左右时间，让 tunnel开始工作。\n",
"\n",
"Set your OpenAI API key below: 随便填写，比如111\n",
"\n",
" \n",
"### 非流式接口\n",
"\n",
"```bash\n",
"curl -vvv https://6d8f-130-211-208-193.ngrok.io/v1/chat/completions \\\n",
" -H \"Content-Type: application/json\" \\\n",
" -H \"Authorization: Bearer token1\" \\\n",
" -d '{ \"stream\": false,\n",
" \"model\": \"gpt-3.5-turbo\",\n",
" \"messages\": [{\"role\": \"user\", \"content\": \"写一首夏天的诗\"}]\n",
" }'\n",
"```\n",
"\n",
"输出：\n",
"```json\n",
"{\"id\":\"chatcmpl-77PZm95TtxE0oYLRx3cxa6HtIDI7s\",\"object\":\"chat.completion\",\"created\":1682000966,\"model\":\"gpt-3.5-turbo-0301\",\"usage\":{\"prompt_tokens\":10,\"completion_tokens\":10,\"total_tokens\":20},\"choices\":[{\"message\":{\"role\":\"assistant\",\"content\":\"夏日的阳光下，\\n树叶闪烁着翠绿的光芒，\\n蝉鸣声不断响起，\\n伴着鸟儿的欢快歌唱。\\n\\n人们穿着轻便的衣服，\\n享受这清凉的夏日时光，\\n在海滩上晒着太阳，\\n喝着清凉的饮料，\\n聊天、欢笑、无忧无虑。\\n\\n清晨的日出，\\n天边呈现出美丽的红色，\\n太阳慢慢地升起，\\n照耀着整个天空。\\n\\n在公园里漫步，\\n欣赏着花草树木的美丽，\\n夏日的天空和大地，\\n让人感到无限的快乐。\\n\\n夏日的风吹过，\\n带来了凉爽的感觉，\\n让人感受到生命的美好，\\n让人感受到夏日的热情。\\n\\n这是一个美好的季节，\\n一个充满欢乐和热情的夏日，\\n让我们珍惜这美好的时光，\\n享受这夏日带来的快乐。\"},\"finish_reason\":\"stop\",\"index\":0}]}\n",
"```\n",
"\n",
"### 流式接口\n",
"\n",
"```bash\n",
"curl -vvv https://6d8f-130-211-208-193.ngrok.io/v1/chat/completions \\\n",
" -H \"Content-Type: application/json\" \\\n",
" -H \"Authorization: Bearer token1\" \\\n",
" -d '{ \"stream\": true, \n",
" \"model\": \"gpt-3.5-turbo\",\n",
" \"messages\": [{\"role\": \"user\", \"content\": \"写一首夏天的诗\"}]\n",
" }'\n",
"```\n",
"\n",
"输出：\n",
"```json\n",
"\n",
"data: {\"id\": \"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\", \"object\": \"chat.completion.chunk\", \"created\": 1682004627, \"model\": \"gpt-3.5-turbo-0301\", \"choices\": [{\"delta\": {\"content\": \"难忘的\"}, \"index\": 0, \"finish_reason\": null}]}\n",
"\n",
"data: {\"id\": \"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\", \"object\": \"chat.completion.chunk\", \"created\": 1682004627, \"model\": \"gpt-3.5-turbo-0301\", \"choices\": [{\"delta\": {\"content\": \"夏日\"}, \"index\": 0, \"finish_reason\": null}]}\n",
"\n",
"data: {\"id\": \"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\", \"object\": \"chat.completion.chunk\", \"created\": 1682004627, \"model\": \"gpt-3.5-turbo-0301\", \"choices\": [{\"delta\": {\"content\": \"时光\"}, \"index\": 0, \"finish_reason\": null}]}\n",
"\n",
"data: {\"id\": \"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\", \"object\": \"chat.completion.chunk\", \"created\": 1682004627, \"model\": \"gpt-3.5-turbo-0301\", \"choices\": [{\"delta\": {\"content\": \"。\"}, \"index\": 0, \"finish_reason\": null}]}\n",
"\n",
"data: {\"id\": \"chatcmpl-77QWpn5cxFi9sVMw56DZReDiGKmcB\", \"object\": \"chat.completion.chunk\", \"created\": 1682004627, \"model\": \"gpt-3.5-turbo-0301\", \"choices\": [{\"delta\": {}, \"index\": 0, \"finish_reason\": \"stop\"}]}\n",
"\n",
"data: [DONE]\n",
"```\n"
],
"metadata": {
"id": "xd6IIdSaIUhr"
}
},
{
"cell_type": "code",
"source": [
"!ps aux | grep cloudflare"
],
"metadata": {
"id": "rJeaEOP9X3-4"
},
"execution_count": null,
"outputs": []
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment