Communicating with Neural Networks Using Telegram

In this article, you will learn how to use ChatGPT (OpenAI), Midjourney, Stable Defusion neural networks using Telegram.

Free text generation with limits

Fast start:

1. Write to the bot directly @ChatGPT_BIArobot
2. Subscribe to the channel @MaxGPT
3. Send a request to @ChatGPT_BIArobot

We use the official open API to communicate with neural networks and we pay for requests, so there are limits for you:

1. Use only English without special characters or arithmetic operators
2. Message must be between 30 and 200 characters
3. Do not use profanity and NSFW content

If you are not satisfied with the limits, read the continuation of this article or register with the service you need to generate text or images.

Generate text and images without limits

Go to @chat_gpt_robot_chat bot and pay the required number of tokens.
1 token ~ 4 characters
from 2000 tokens = 1 request for image (depends on the model)

Text generation

Before generating text, change the bot’s operation mode to text generation by sending the /text command.

If you want to use a specific text generation model, send the /settings command to the bot, then select “Change model settings” and select the desired model. We support 4 models (text-davinci-003, text-curie-001 etc) for communication and 2 models (code-davinci-002 and code-cushman-001) to generate text. You can read more about them in the bot in the Settings section.

You can change the hard limit on the number of tokens in the request and response using the “Change maximum characters” item. Be careful if you set it too high and if your request is big enough the bot won’t have enough tokens to complete the response. We recommend setting 2000 characters.

Image Generation

Before generating images, change the bot’s operation mode to generate images by sending the /image command.

If you want to use a specific model, send the /settings command to the bot, select “Change image model” and you will be prompted to select a model. The response provides an example of the result of the work, the name and description of the model, the cost of generation, as well as the approximate generation time.