• Log in with your Poe tokens
• Auto Proxy requests
• Specify Proxy context

• Create new chat thread
• Send messages
• Stream bot responses
• Send concurrent messages
• Retry the last message
• Support file attachments
• Retrieve suggested replies
• Stop message generation
• Delete chat threads
• Clear conversation context
• Purge messages of 1 bot
• Purge all messages of user
• Fetch previous messages
• Share and import messages
• Get citations

• Get Chat Ids & Chat Codes of bot(s)
• Get subscription info and remaining points

• Get bot info
• Get available creation models
• Create custom bot
• Edit custom bot
• Delete a custom bot

• Get available knowledge bases
• Upload knowledge bases for custom bots
• Edit knowledge bases for custom bots

• Get available bots
• Get a user's bots
• Get available categories
• Explore 3rd party bots and users

• Create a group chat
• Delete a group chat
• Get created groups
• Get group data
• Save group chat history
• Load group chat history

Available Routes

• /models
• /chat/completions
• /images/generations
• /images/edits
• /v1/models
• /v1/chat/completions
• /v1/images/generations
• /v1/images/edits

Quick Setup

• First, install the additional packages:

pip install -U 'poe-api-wrapper[llm]'

• Clone the repo or use the same setup in openai folder:

git clone https://github.com/snowby666/poe-api-wrapper.git
cd poe-api-wrapper\poe_api_wrapper\openai

• Modify secrets.json with your own tokens

• Run the FastAPI server:

python api.py

• Run the examples:

python example.py

Built-in completion (WIP)

OpenAI Proxy Server

• Start the server

from poe_api_wrapper import PoeServer
tokens = [
    {"p-b": "XXXXXXXX", "p-lat": "XXXXXXXX"},
    {"p-b": "XXXXXXXX", "p-lat": "XXXXXXXX"},
    {"p-b": "XXXXXXXX", "p-lat": "XXXXXXXX"}
]
PoeServer(tokens=tokens)

# You can also specify address and port (default is 127.0.0.1:8000)
PoeServer(tokens=tokens, address="0.0.0.0", port="8080")

Chat

• Non-streamed example:

import openai 
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

response = client.chat.completions.create(
    model="gpt-3.5-turbo", 
    messages = [
                {"role": "system", "content": "You are a helpful assistant."},
                {"role": "user", "content": "Hello!"}
            ]
)

print(response.choices[0].message.content)

• Streaming example:

import openai 
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

stream = client.chat.completions.create(
    model="gpt-3.5-turbo", 
    messages = [
                {"role": "user", "content": "this is a test request, write a short poem"}
            ],
    stream=True
)

for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="", flush=True)

# Set max_tokens
stream_2 = client.chat.completions.create(
    model="claude-instant", 
    messages = [
                {"role": "user", "content": "Can you tell me about the creation of blackholes?"}
            ],
    stream=True,
    max_tokens=20, # if max_tokens reached, finish_reason will be 'length'
)

for chunk in stream_2:
    print(chunk.choices[0].delta.content or "", end="", flush=True)

# Include usage 
stream_3 = client.chat.completions.create(
    model="claude-instant", 
    messages = [
                {"role": "user", "content": "Write a 100-character meta description for my blog post about llamas"}
            ],
    stream=True,
    max_tokens=4096,
    stream_options={
		"include_usage": True # last chunk contains prompts_tokens, completion_tokens and total_tokens
	}
)

for chunk in stream_3:
    print(chunk, end="\n\n", flush=True)

• Image input example:

import openai 
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

# Legacy style (https://platform.openai.com/docs/api-reference/chat/create)
response = client.chat.completions.create(
    model="claude-3.5-sonnet",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {
                    "type": "image_url",
                    "image_url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg",
                }
            ],
        }
    ]
)

# New style (https://platform.openai.com/docs/guides/vision)
response = client.chat.completions.create(
    model="claude-3.5-sonnet",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg"
                    }
                }
            ],
        }
    ]
)

# Multiple images
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "What are in these images? Is there any difference between them?",
        },
        {
          "type": "image_url",
          "image_url": {
            "url": "https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg",
          },
        },
        {
          "type": "image_url",
          "image_url": {
            "url": "https://imgcdn.stablediffusionweb.com/2024/4/29/0b0b8798-1965-4e3d-b0a8-d153728320d4.jpg",
          }
        }
      ]
    }
  ]
)

# Base64 image
import base64

# Function to encode the image
def encode_image(image_path):
  with open(image_path, "rb") as image_file:
    return base64.b64encode(image_file.read()).decode('utf-8')

# Path to your image
image_path = "path_to_your_image.jpg"

# Getting the base64 string
base64_image = encode_image(image_path)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "What’s in this image?"
        },
        {
          "type": "image_url",
          "image_url": {
            "url": f"data:image/jpeg;base64,{base64_image}"
          }
        }
      ]
    }
  ]
)

print(response.choices[0].message.content)

• Function calling example:

import openai, json
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

TEST_MODEL = "gpt-4o-mini"

# Example dummy function hard coded to return the same weather
# In production, this could be your backend API or an external API
def get_current_temperature(location, unit="fahrenheit"):
    """Get the current weather in a given location"""
    if "tokyo" in location.lower():
        return json.dumps({"location": "Tokyo", "temperature": "10", "unit": unit})
    elif "san francisco" in location.lower():
        return json.dumps({"location": "San Francisco", "temperature": "72", "unit": unit})
    elif "paris" in location.lower():
        return json.dumps({"location": "Paris", "temperature": "22", "unit": unit})
    else:
        return json.dumps({"location": location, "temperature": "unknown"})
    
def get_rain_probability(location):
    """Get the probability of rain in a given location"""
    if "tokyo" in location.lower():
        return json.dumps({"location": "Tokyo", "rain_probability": "10%"})
    elif "san francisco" in location.lower():
        return json.dumps({"location": "San Francisco", "rain_probability": "20%"})
    elif "paris" in location.lower():
        return json.dumps({"location": "Paris", "rain_probability": "30%"})
    else:
        return json.dumps({"location": location, "rain_probability": "unknown"})
    
def run_conversation():
    # Step 1: send the conversation and available functions to the model
    messages = [
        {'role': 'user', 'content': "Hello there. What the weather like in Tokyo?"},
        {'role': 'assistant', 'content': "Let me check the weather for you."},
        {'role': 'user', 'content': "What is the chance of raining in paris? Can you also tell me the temperature in Tokyo and LA?"},
                ]
    tools = [
    {
      "type": "function",
      "function": {
        "name": "get_current_temperature",
        "description": "Get the current temperature for a specific location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "The city and state, e.g., San Francisco, CA"
            },
            "unit": {
              "type": "string",
              "enum": ["Celsius", "Fahrenheit"],
              "description": "The temperature unit to use. Infer this from the user's location."
            }
          },
          "required": ["location", "unit"]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "get_rain_probability",
        "description": "Get the probability of rain for a specific location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "The city and state, e.g., San Francisco, CA"
            }
          },
          "required": ["location"]
        }
      }
    }
  ]
    response = client.chat.completions.create(
        model=TEST_MODEL,
        messages=messages,
        tools=tools,
        tool_choice={"type": "function", "function": {"name": "get_current_temperature"}},
    )
    response_message = response.choices[0].message
    print("\n", response_message, "\n")
        
    tool_calls = response_message.tool_calls
    # Step 2: check if the model wanted to call a function
    if tool_calls:
        # Step 3: call the function
        # Note: the JSON response may not always be valid; be sure to handle errors
        available_functions = {
            "get_current_temperature": get_current_temperature,
            "get_rain_probability": get_rain_probability
        }  # only two functions in this example, but you can have multiple
        messages.append(response_message)  # extend conversation with assistant's reply
        # Step 4: send the info for each function call and function response to the model
        for tool_call in tool_calls:
            print(tool_call, "\n")
            function_name = tool_call.function.name
            function_to_call = available_functions[function_name]
            function_args = json.loads(tool_call.function.arguments)
            function_response = function_to_call(**function_args)
            messages.append(
                {
                    "tool_call_id": tool_call.id,
                    "role": "tool",
                    "name": function_name,
                    "content": function_response,
                }
            )  # extend conversation with function response
    second_response = client.chat.completions.create(
        model=TEST_MODEL,
        messages=messages,
    )  # get a new response from the model where it can see the function response
    return second_response.choices[0].message.content

print(run_conversation())

Images

• Create image example:

import openai
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

images_url = client.images.generate(
  model="playground-v2.5",
  prompt="A cute baby sea otter",
  n=2, # The number of images to generate
  size="1792x1024" # The size of image (view models.json for available sizes)
)

print(images_url)

• Edit image example:

import openai
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

images_url = client.images.edit(
  image="https://imgcdn.stablediffusionweb.com/2024/4/29/0b0b8798-1965-4e3d-b0a8-d153728320d4.jpg",
  model="sdxl",
  prompt="A cute baby sea otter wearing a raincoat",
  n=1, # The number of images to generate
  size="1024x1024" # The size of image (view models.json for available sizes)
)

print(images_url)

Models

• List models example:

import openai
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

models = client.models.list()

print(models)

• Retrieve model example:

import openai
client = openai.OpenAI(api_key="anything", base_url="http://127.0.0.1:8000/v1/", default_headers={"Authorization": "Bearer anything"})

model = client.models.retrieve("gpt-3.5-turbo-instruct")

print(model)

• Connecting to the API

tokens = {
    'p-b': 'p-b cookie here',
    'p-lat': 'p-lat cookie here',
}

# Default setup
from poe_api_wrapper import PoeApi
client = PoeApi(tokens=tokens)

# Using Client with auto_proxy (default is False)
client = PoeApi(tokens=tokens, auto_proxy=True)

# Passing proxies manually
proxy_context = [
    {"https://":X1, "http://":X1},
    {"https://":X2, "http://":X2},
    ...
]

client = PoeApi(tokens=tokens, proxy=proxy_context) 

# Add formkey and cloudflare cookies to pass challenges
tokens = {
    'p-b': 'p-b cookie here',
    'p-lat': 'p-lat cookie here',
    'formkey': 'formkey here',
    '__cf_bm': '__cf_bm cookie here', 
    'cf_clearance': 'cf_clearance cookie here'
}

• Getting Chat Ids & Chat Codes

# Get chat data of all bots (this will fetch all available threads)
print(client.get_chat_history()['data'])
>> Output:
{'chinchilla': [{'chatId': 74397929, 'chatCode': '2ith0h11zfyvsta1u3z', 'id': 'Q2hhdDo3NDM5NzkyOQ==', 'title': 'Comparison'}], 'code_llama_7b_instruct': [{'chatId': 74397392, 'chatCode': '2ithbduzsysy3g178hb', 'id': 'Q2hhdDo3NDM5NzM5Mg==', 'title': 'Decent Programmers'}], 'a2': [{'chatId': 74396838, 'chatCode': '2ith9nikybn4ksn51l8', 'id': 'Q2hhdDo3NDM5NjgzOA==', 'title': 'Reverse Engineering'}, {'chatId': 74396452, 'chatCode': '2ith79n4x0p0p8w5yue', 'id': 'Q2hhdDo3NDM5NjQ1Mg==', 'title': 'Clean Code'}], 'leocooks': [{'chatId': 74396246, 'chatCode': '2ith82wj0tjrggj46no', 'id': 'Q2hhdDo3NDM5NjI0Ng==', 'title': 'Pizza perfection'}], 'capybara': [{'chatId': 74396020, 'chatCode': '2ith5o3p8c5ajkdwd3k', 'id': 'Q2hhdDo3NDM5NjAyMA==', 'title': 'Greeting'}]}

# Get chat data of a bot (this will fetch all available threads)
print(client.get_chat_history("a2")['data'])
>> Output:
{'a2': [{'chatId': 74396838, 'chatCode': '2ith9nikybn4ksn51l8', 'id': 'Q2hhdDo3NDM5NjgzOA==', 'title': 'Reverse Engineering'}, {'chatId': 74396452, 'chatCode': '2ith79n4x0p0p8w5yue', 'id': 'Q2hhdDo3NDM5NjQ1Mg==', 'title': 'Clean Code'}]}

# Get a defined number of most recent chat threads (using count param will ignore interval param)
# Fetching all bots
print(client.get_chat_history(count=20)['data'])
# Fetching 1 bot
print(client.get_chat_history(bot="a2", count=20)['data'])

# You can pass the number of bots fetched for each interval to both functions. (default is 50)
# Fetching 200 chat threads of all bots each interval
print(client.get_chat_history(interval=200)['data'])
# Fetching 200 chat threads of a bot each interval
print(client.get_chat_history(bot="a2", interval=200)['data'])

# Pagination Example:
# Fetch the first 20 chat threads
history = client.get_chat_history(count=20)
pages = [history['data']]
new_cursor = history['cursor']

# Set a while loop with a condition of your choice
while new_cursor != None:
    # Fetch the next 20 chat threads with new_cursor
    new_history = client.get_chat_history(count=20, cursor=new_cursor)
    # Append the next 20 chat threads 
    new_cursor = new_history['cursor']
    pages.append(new_history['data'])

# Print the pages (20 chat threads each page)
for page in range(len(pages)):
    print(f'This is page {page+1}')
    for bot, value in pages[page].items():
        for thread in value:
            print({bot: thread})

• Getting subscription info and remaining points

data = client.get_settings()
print(data)

• Sending messages & Streaming responses

bot = "a2"
message = "What is reverse engineering?"

# Create new chat thread
# Streamed example:
for chunk in client.send_message(bot, message):
    print(chunk["response"], end="", flush=True)
print("\n")

# Non-streamed example:
for chunk in client.send_message(bot, message):
    pass
print(chunk["text"])

# You can get chatCode and chatId of created thread to continue the conversation
chatCode = chunk["chatCode"]
chatId = chunk["chatId"]
# You can also retrieve msgPrice
msgPrice = chunk["msgPrice"]

# Send message to an existing chat thread
# 1. Using chatCode
for chunk in client.send_message(bot, message, chatCode="2i58ciex72dom7im83r"):
    print(chunk["response"], end="", flush=True)
# 2. Using chatId
for chunk in client.send_message(bot, message, chatId=59726162):
    print(chunk["response"], end="", flush=True)
# 3. Specify msgPrice manually (the wrapper automatically gets this, but you can also pass the param for less resources consumed)
for chunk in client.send_message(bot, message, chatId=59726162, msgPrice=msgPrice):
    print(chunk["response"], end="", flush=True)

> [!NOTE] > Display names are the same as the codenames for custom bots, you can simply pass the bot's display name into client.send_message(bot, message)

• Sending concurrent messages

# Use at your own risk, increase timeout to avoid ratelimit (default is 20)

import time, threading
thread_count = 0

def message_thread(prompt, counter):
    global thread_count
    try:
        for chunk in client.send_message("gpt3_5", prompt):
            pass
        print(prompt+"\n"+chunk["text"]+"\n"*3)
        thread_count -= 1
    except Exception as e:
        pass

prompts = [
  "Write a paragraph about the impact of social media on mental health.",
  "Write a paragraph about the history and significance of the Olympic Games.",
  "Write a paragraph about the effects of climate change on the world's oceans.",
  "Write a paragraph about the benefits and drawbacks of remote work for employees and companies.",
  "Write a paragraph about the role of technology in modern education.",
  "Write a paragraph about the history and impact of the Civil Rights Movement in America.",
  "Write a paragraph about the impact of COVID-19 on global economies.",
  "Write a paragraph about the rise and fall of the Roman Empire.",
  "Write a paragraph about the benefits and drawbacks of genetically modified organisms (GMOs).",
  "Write a paragraph about the impact of globalization on cultural identity.",
  "Write a paragraph about the history and significance of the Mona Lisa painting.",
  "Write a paragraph about the benefits and drawbacks of renewable energy sources.",
  "Write a paragraph about the impact of social media on political discourse.",
  "Write a paragraph about the history and impact of the Industrial Revolution.",
  "Write a paragraph about the benefits and drawbacks of online shopping for consumers and businesses.",
  "Write a paragraph about the impact of artificial intelligence on the job market.",
  "Write a paragraph about the history and significance of the Great Wall of China.",
  "Write a paragraph about the benefits and drawbacks of standardized testing in schools.",
  "Write a paragraph about the impact of the feminist movement on women's rights.",
  "Write a paragraph about the history and impact of the American Revolution."
]

   
for i in range(len(prompts)):
    t = threading.Thread(target=message_thread, args=(prompts[i], i), daemon=True)
    t.start()
    thread_count += 1
    time.sleep(1)

while thread_count:
    time.sleep(0.01)

• Retrying the last message

for chunk in client.retry_message(chatCode):
    print(chunk['response'], end='', flush=True)

• Adding file attachments

# Web urls example:
file_urls = ["https://elinux.org/images/c/c5/IntroductionToReverseEngineering_Anderson.pdf", 
            "https://www.kcl.ac.uk/warstudies/assets/automation-and-artificial-intelligence.pdf"]
for chunk in client.send_message(bot, "Compare 2 files and describe them in 300 words", file_path=file_urls):
    print(chunk["response"], end="", flush=True)
    
# Local paths example:
local_paths = ["c:\\users\\snowby666\\hello_world.py"]
for chunk in client.send_message(bot, "What is this file about?", file_path=local_paths):
    print(chunk["response"], end="", flush=True)

> [!NOTE] > The files size limit is different for each model.

• Retrieving suggested replies

for chunk in client.send_message(bot, "Introduce 5 books about clean code", suggest_replies=True):
    print(chunk["response"], end="", flush=True)
print("\n")

for reply in chunk["suggestedReplies"]:
    print(reply)

• Stopping message generation

# You can use an event to trigger this function
# Example:
# Note that keyboard library may not be compatible with MacOS, Linux, Ubuntu
import keyboard
for chunk in client.send_message(bot, message):
    print(chunk["response"], end="", flush=True)
    # Press Q key to stop the generation
    if keyboard.is_pressed('q'):
        client.cancel_message(chunk)
        print("\nMessage is now cancelled")
        break 

• Deleting chat threads

# Delete 1 chat
# Using chatCode
client.delete_chat(bot, chatCode="2i58ciex72dom7im83r")
# Using chatId
client.delete_chat(bot, chatId=59726162)

# Delete n chats
# Using chatCode
client.delete_chat(bot, chatCode=["LIST_OF_CHAT_CODES"])
# Using chatId
client.delete_chat(bot, chatId=["LIST_OF_CHAT_IDS"])

# Delete all chats of a bot
client.delete_chat(bot, del_all=True)

• Clearing conversation context

# 1. Using chatCode
client.chat_break(bot, chatCode="2i58ciex72dom7im83r")
# 2. Using chatId
client.chat_break(bot, chatId=59726162)

• Purging messages of 1 bot

# Purge a defined number of messages (default is 50)
# 1. Using chatCode
client.purge_conversation(bot, chatCode="2i58ciex72dom7im83r", count=10)
# 2. Using chatId
client.purge_conversation(bot, chatId=59726162, count=10)

# Purge all messsages of the thread
# 1. Using chatCode
client.purge_conversation(bot, chatCode="2i58ciex72dom7im83r", del_all=True)
# 2. Using chatId
client.purge_conversation(bot, chatId=59726162,  del_all=True)

• Purging all messages of user

client.purge_all_conversations()

• Fetching previous messsages

# Get a defined number of messages (default is 50)
# Using chatCode
previous_messages = client.get_previous_messages('code_llama_34b_instruct', chatCode='2itg2a7muygs42v1u0k', count=2)
# Using chatId
previous_messages = client.get_previous_messages('code_llama_34b_instruct', chatId=74411139, count=2)
for message in previous_messages:
    print(message)
>> Output:
{'author': 'human', 'text': 'nice to meet you', 'messageId': 2861709279}
{'author': 'code_llama_34b_instruct', 'text': " Nice to meet you too! How are you doing today? Is there anything on your mind that you'd like to talk about? I'm here to listen and help", 'messageId': 2861873125}

# Get all previous messages
# Using chatCode
previous_messages = client.get_previous_messages('code_llama_34b_instruct', chatCode='2itg2a7muygs42v1u0k', get_all=True)
# Using chatId
previous_messages = client.get_previous_messages('code_llama_34b_instruct', chatId=74411139, get_all=True)
for message in previous_messages:
    print(message)
>> Output:
{'author': 'human', 'text': 'hi there', 'messageId': 2861363514}
{'author': 'code_llama_34b_instruct', 'text': " Hello! It's nice to meet you. Is there something I can help you with or would you like to chat?", 'messageId': 2861363530}
{'author': 'chat_break', 'text': "", 'messageId': 2872383991}
{'author': 'human', 'text': 'nice to meet you', 'messageId': 2861709279}
{'author': 'code_llama_34b_instruct', 'text': " Nice to meet you too! How are you doing today? Is there anything on your mind that you'd like to talk about? I'm here to listen and help", 'messageId': 2861873125}

> [!NOTE] > It will fetch messages from the latest to the oldest, but the order to be displayed is reversed.

• Getting available knowledge bases

# Get a defined number of sources (default is 10)
print(client.get_available_knowledge(botName="BOT_NAME", count=2))
>> Output:
{'What is Quora?': [86698], 'Founders of Quora': [86705]}
# Get all available sources
print(client.get_available_knowledge(botName="BOT_NAME", get_all=True))

• Uploading knowledge bases

# Web urls example:
file_urls = ["https://elinux.org/images/c/c5/IntroductionToReverseEngineering_Anderson.pdf", 
            "https://www.kcl.ac.uk/warstudies/assets/automation-and-artificial-intelligence.pdf"]
source_ids = client.upload_knowledge(file_path=file_urls)
print(source_ids)
>> Output:
{'er-1-intro_to_re.pdf': [86344], 'automation-and-artificial-intelligence.pdf': [86345]}

# Local paths example:
local_paths = ["c:\\users\\snowby666\\hello_world.py"]
source_ids = client.upload_knowledge(file_path=local_paths)
print(source_ids)
>> Output:
{'hello_world.py': [86523]}

# Plain texts example:
knowledges = [
    {
        "title": "What is Quora?",
        "content": "Quora is a popular online platform that enables users to ask questions on various topics and receive answers from a diverse community. It covers a wide range of subjects, from academic and professional queries to personal experiences and opinions, fostering knowledge-sharing and meaningful discussions among its users worldwide."
    },
    {
        "title": "Founders of Quora",
        "content": "Quora was founded by two individuals, Adam D'Angelo and Charlie Cheever. Adam D'Angelo, who previously served as the Chief Technology Officer (CTO) at Facebook, and Charlie Cheever, a former Facebook employee as well, launched Quora in June 2009. They aimed to create a platform that would enable users to ask questions and receive high-quality answers from knowledgeable individuals. Since its inception, Quora has grown into a widely used question-and-answer platform with a large user base and a diverse range of topics covered."
    },
]
source_ids = client.upload_knowledge(text_knowledge=knowledges)
print(source_ids)
>> Output:
{'What is Quora?': [86368], 'Founders of Quora': [86369]}

# Hybrid example:
source_ids = client.upload_knowledge(file_path=file_urls, text_knowledge=knowledges)
print(source_ids)
>> Output:
{'What is Quora?': [86381], 'Founders of Quora': [86383], 'er-1-intro_to_re.pdf': [86395], 'automation-and-artificial-intelligence.pdf': [86396]}

• Editing knowledge bases (Only for plain texts)

client.edit_knowledge(knowledgeSourceId=86381, title='What is Quora?', content='Quora is a question-and-answer platform where users can ask questions, provide answers, and engage in discussions on various topics.')

• Getting bot info

bot = 'gpt-4'
print(client.get_botInfo(handle=bot))
>> Output:
{'handle': 'GPT-4', 'model': 'beaver', 'supportsFileUpload': True, 'messageTimeoutSecs': 15, 'displayMessagePointPrice': 350, 'numRemainingMessages': 20, 'viewerIsCreator': False, 'id': 'Qm90OjMwMDc='}

• Getting available creation models

print(client.get_available_creation_models())
>> Output:
{'text': ['claude_3_igloo', 'gpt4_o_mini', 'gpt4_o', 'gemini_1_5_flash', 'gemini_1_5_pro', 'claude_2_1_bamboo', 'claude_3_haiku', 'claude_2_1_cedar', 'gemini_1_5_flash_128k', 'gemini_1_5_pro_128k', 'gemini_1_5_flash_1m', 'gemini_1_5_pro_1m', 'gpt4_o_mini_128k', 'gpt4_o_128k', 'beaver', 'gemini_pro', 'chinchilla', 'vizcacha', 'claude_3_igloo_200k', 'claude_3_sonnet_200k', 'claude_3_haiku_200k', 'claude_3_opus_200k', 'mixtral8x7bchat', 'claude_2_short', 'a2_2', 'mythomaxl213b', 'a2', 'a2_100k'], 'image': ['playgroundv25', 'ideogram', 'dalle3', 'stablediffusion3', 'sd3turbo', 'stablediffusionxl'], 'video': ['pika']}

• Creating a new Bot

client.create_bot(handle="BOT_NAME", prompt="PROMPT_HERE", base_model="a2")

# Using knowledge bases (you can use source_ids from uploaded knowledge bases for your custom bot)
client.create_bot(handle="BOT_NAME", prompt="PROMPT_HERE", base_model="a2", knowledgeSourceIds=source_ids, shouldCiteSources=True)

• Editing a Bot

client.edit_bot(handle="BOT_NAME", prompt="PROMPT_HERE", new_handle="NEW_BOT_NAME", base_model='chinchilla')

# Adding knowledge bases 
client.edit_bot(handle="BOT_NAME", prompt="PROMPT_HERE", new_handle="NEW_BOT_NAME", base_model='chinchilla', knowledgeSourceIdsToAdd=source_ids, shouldCiteSources=True)

# Removing knowledge bases
client.edit_bot(handle="BOT_NAME", prompt="PROMPT_HERE", new_handle="NEW_BOT_NAME", base_model='chinchilla', knowledgeSourceIdsToRemove=source_ids, shouldCiteSources=True)

> [!TIP] > You can also use both knowledgeSourceIdsToAdd and knowledgeSourceIdsToRemove at the same time.

• Deleting a Bot

client.delete_bot(handle="BOT_NAME")

• Getting available bots (your bots section)

# Get a defined number of bots (default is 25)
print(client.get_available_bots(count=10))
# Get all available bots
print(client.get_available_bots(get_all=True))

• Getting a user's bots

handle = 'poe'
print(client.get_user_bots(user=handle))

• Getting available categories

print(client.get_available_categories())
>> Output:
['Official', 'Popular', 'New', 'ImageGen', 'AI', 'Professional', 'Funny', 'History', 'Cooking', 'Advice', 'Mind', 'Programming', 'Travel', 'Writing', 'Games', 'Learning', 'Roleplay', 'Utilities', 'Sports', 'Music']

• Exploring 3rd party bots and users

# Explore section example:
# Get a defined number of bots (default is 50)
print(client.explore(count=10))
# Get all available bots
print(client.explore(explore_all=True))

# Search for bots by query example:
# Get a defined number of bots (default is 50)
print(client.explore(search="Midjourney", count=30))
# Get all available bots
print(client.explore(search="Midjourney", explore_all=True))

# Search for bots by category example (default is defaultCategory):
# Get a defined number of bots (default is 50)
print(client.explore(categoryName="Popular", count=30))
# Get all available bots
print(client.explore(categoryName="AI", explore_all=True))

# Search for people example:
# Get a defined number of people (default is 50)
print(client.explore(search="Poe", entity_type='user', count=30))
# Get all available people
print(client.explore(search="Poe", entity_type='user', explore_all=True))

• Sharing & Importing messages

# Share a defined number of messages (from the lastest to the oldest)
# Using chatCode
shareCode = client.share_chat("a2", chatCode="2roap5g8nd7s28ul836",count=10)
# Using chatId
shareCode = client.share_chat("a2", chatId=204052028,count=10)

# Share all messages
# Using chatCode
shareCode = client.share_chat("a2", chatCode="2roap5g8nd7s28ul836")
# Using chatId
shareCode = client.share_chat("a2", chatId=204052028)

# Set up the 2nd Client and import messages from the shareCode
client2 = PoeApi("2nd_TOKEN_HERE")
print(client2.import_chat(bot, shareCode))
>> Output:
{'chatId': 72929127, 'chatCode': '2iw0xcem7a18wy1avd3'}

• Getting citations

print(client.get_citations(messageId=141597902621))

• Creating a group chat

bots = [
    {'bot': 'yayayayaeclaude', 'name': 'Yae'}, 
    {'bot': 'gepardL', 'name': 'gepard'}, 
    {'bot': 'SayukiTokihara', 'name': 'Sayuki'}
]

client.create_group(group_name='Hangout', bots=bots) 

> [!NOTE] > bot arg is the model/displayName. > name arg is the one you'd mention them in group chat.

• Sending messages and Streaming responses in group chat

# User engagement example:
while True: 
    message = str(input('\n\033[38;5;121mYou : \033[0m'))
    prev_bot = ""
    for chunk in client.send_message_to_group(group_name='Hangout', message=message):
        if chunk['bot'] != prev_bot:
            print(f"\n\033[38;5;121m{chunk['bot']} : \033[0m", end='', flush=True)
            prev_bot = chunk['bot']
        print(chunk['response'], end='', flush=True)
    print('\n')

# Auto-play example:
while True:
    prev_bot = ""
    for chunk in client.send_message_to_group(group_name='Hangout', autoplay=True):
        if chunk['bot'] != prev_bot:
            print(f"\n\033[38;5;121m{chunk['bot']} : \033[0m", end='', flush=True)
            prev_bot = chunk['bot']
        print(chunk['response'], end='', flush=True)
    print('\n')

# Preset history example:
preset_path = "c:\\users\\snowby666\\preset.json"
prev_bot = ""
for chunk in client.send_message_to_group(group_name='Hangout', autoplay=True, preset_history=preset_path):
    if chunk['bot'] != prev_bot:
        print(f"\n\033[38;5;121m{chunk['bot']} : \033[0m", end='', flush=True)
        prev_bot = chunk['bot']
    print(chunk['response'], end='', flush=True)
print('\n')
while True:
    for chunk in client.send_message_to_group(group_name='Hangout', autoplay=True):
        if chunk['bot'] != prev_bot:
            print(f"\n\033[38;5;121m{chunk['bot']} : \033[0m", end='', flush=True)
            prev_bot = chunk['bot']
        print(chunk['response'], end='', flush=True)
    print('\n')

> [!NOTE] > You can also change your name in group chat by passing a new one to the above function: client.send_message_to_group('Hangout', message=message, user='Danny') > If you want to auto save the conversation_log, just simply set this to true: client.send_message_to_group('Hangout', message=message, autosave=True)

• Deleting a group chat

client.delete_group(group_name='Hangout')

• Getting created groups

print(client.get_available_groups())

• Getting group data

print(client.get_group(group_name='Hangout'))

• Saving group chat history

# Save as json in the same directory
client.save_group_history(group_name='Hangout')
# Save with a local path (json only)
local_path = "c:\\users\\snowby666\\log.json"
client.save_group_history(group_name='Hangout', file_path=local_path)

• Loading group chat history

print(client.load_group_history(file_path=local_path))

• How to find chatCode manually?

Here is an example, the chatCode is 23o1gxjhb9cfnlacdcd

• What are the file types that poe-api-wrapper support?

Currently, this API only supports these file types for adding attachments

Text files

Media files