Welcome to the Deep.Assistant organization! We are dedicated to building AI-powered applications that enhance user experiences across various platforms. Our current projects include:

• api-gateway: A central API gateway managing backend services for our applications and direct API users.
• telegram-bot: A Telegram bot (Deep.GPT) providing AI-powered assistance.
• GPTutor: A VK mini-app for AI-driven tutoring.

The following diagram illustrates how our applications interact with each other and the backend services. Users can also directly access the API gateway for OpenAI-compatible services. Click on the components to view their detailed architecture documentation.

graph TB
subgraph UserLayer["User Layer"]
direct_api_users["Direct API Users"]
user_tele["Telegram Users"]
user_vk["VK Users"]
end
subgraph ApplicationLayer["Application Layer"]
app_telebot["telegram-bot (Python)"]
app_gptutor["GPTutor (React, Typescript)"]
end
subgraph MiddlewareLayer["Middleware Layer"]
middleware_apigw["api-gateway (Node.js, JavaScript)"]
end
subgraph backendLayer["Backend Layer"]
backend_ai["AI API Services"]
backend_db["Database"]
end
direct_api_users <--> middleware_apigw
user_tele <--> app_telebot
user_vk <--> app_gptutor
app_telebot <--> middleware_apigw
app_gptutor <--> middleware_apigw
middleware_apigw <--> backend_ai
middleware_apigw <--> backend_db
click app_gptutor "https://github.com/deep-assistant/GPTutor/blob/main/ARCHITECTURE.md"
click app_telebot "https://github.com/deep-assistant/telegram-bot/blob/main/ARCHITECTURE.md"
click middleware_apigw "https://github.com/deep-assistant/api-gateway/blob/main/ARCHITECTURE.md"

• Telegram Bot (Deep.GPT): Start chatting with our AI-powered assistant at https://t.me/DeepGPTBot. Use /help for a list of commands or /api to get your API key.
• GPTutor: Access our AI tutoring mini-app through our VK community at http://vk.com/gptutor. Launch the app directly from the community page.

• API Gateway: Obtain an API key from https://t.me/DeepGPTBot using /api, then follow the quick start guide below to integrate with our OpenAI-compatible API at https://api.deep.assistant.run.place/v1/.
• Contributing: Clone any repository, set it up using the README instructions, and submit a pull request. See Contributing Guidelines below.

• Description: A central API gateway managing requests from GPTutor, telegram-bot, and direct users to various API providers (e.g., GPT, LLM, Image, Music). It's OpenAI-compatible and accessible at https://api.deep.assistant.run.place/v1/.

• Technology: Built with Node.js and Express.js for routing.

• Setup: See the api-gateway README.

• Quick Start:

  JavaScript:

  importOpenAIfrom'openai';importdotenvfrom'dotenv';// Load environment variables from .env filedotenv.config();constOPENAI_API_KEY=process.env.OPENAI_API_KEY||"YOUR_API_KEY";// Get from /api command at https://t.me/DeepGPTBotconstOPENAI_API_BASE=process.env.OPENAI_BASE_URL||process.env.OPENAI_API_BASE||"https://api.deep.assistant.run.place/v1/";constopenai=newOpenAI({apiKey: OPENAI_API_KEY,baseURL: OPENAI_API_BASE});asyncfunctionmain(){constchatCompletion=awaitopenai.chat.completions.create({messages: [{role: 'user',content: 'Hi'}],model: 'gpt-4o-mini',});console.log(chatCompletion.choices[0].message.content);}main();

  → Complete JavaScript Example - Full setup with package.json, dependencies, and instructions

  Python:

  importosfromopenaiimportOpenAIfromdotenvimportload_dotenv# Load environment variables from .env fileload_dotenv()
  OPENAI_API_KEY=os.getenv("OPENAI_API_KEY", "YOUR_API_KEY") # Get from /api command at https://t.me/DeepGPTBotOPENAI_API_BASE=os.getenv("OPENAI_BASE_URL") oros.getenv("OPENAI_API_BASE", "https://api.deep.assistant.run.place/v1/")
  client=OpenAI(
  api_key=OPENAI_API_KEY,
  base_url=OPENAI_API_BASE
  )
  defmain():
  chat_completion=client.chat.completions.create(
  messages=[{"role": "user", "content": "Hi"}],
  model="gpt-4o-mini",
  )
  print(chat_completion.choices[0].message.content)
  if__name__=="__main__":
  main()

  → Complete Python Example - Full setup with requirements.txt, dependencies, and instructions

• API Documentation: See the API Gateway Docs for advanced usage (e.g., streaming, Whisper transcription).

• Description: Branded as Deep.GPT, this Telegram bot offers AI-driven assistance. Start chatting at https://t.me/DeepGPTBot.
• Technology: Developed in Python using the python-Telegram-Bot library.
• Setup: Follow the telegram-bot README.
• User Guide:
  • Start: Send /start to begin.
  • Help: Use /help for commands.
  • API Key: Get yours with /api.
  • Chat: Simply type your questions or prompts!

• Description: A VK mini-app branded as GPTutor, offering AI-powered tutoring, particularly for Russian-speaking users. Access it at http://vk.com/gptutor.
• Technology: Built with React and Typescript, integrated with vk-mini-apps-API.
• Setup: See the GPTutor README.
• User Guide:
  • Access: Visit http://vk.com/gptutor and click the app link.
  • Use: Follow in-app prompts to ask questions or start tutoring sessions.

Below is a list of available AI models supported by Deep.Assistant applications and the API gateway. Pricing is provided per 1K tokens and per 1M tokens, with availability indicated for VK (GPTutor), Telegram (Deep.GPT), and API usage.

Note: 1K⚡️ ≈ 0.8 RUB ≈ 0.01 USD.

• Notes: Costs are based on token usage. "+" indicates availability; "-" indicates unavailability (subject to change).
• Updates: For the latest on model availability or pricing, reach out via Telegram bot at https://t.me/DeepGPTBot.

We welcome contributions! Steps:

1. Fork the repository.
2. Follow setup instructions in the README.
3. Submit a pull request with your changes.

All projects are licensed under the Unlicense, granting maximum freedom to use and modify the code. Verify the license file in each repository.