Skip to content

oomol-flows/oomol-llm

Repository files navigation

LLM - AI Language Model Blocks for OOMOL

A collection of ready-to-use AI language model building blocks for the OOMOL platform, designed to help you create intelligent conversational workflows without writing code.

What is This Project?

This project provides pre-built blocks (reusable components) that allow you to integrate AI language models into your OOMOL workflows. Think of blocks as LEGO pieces - you can connect them together to build powerful AI applications like chatbots, translators, content generators, and more.

Who is This For?

  • Business users who want to automate AI-powered tasks
  • Content creators looking to leverage AI for text generation
  • Product managers prototyping AI features
  • Anyone interested in building AI workflows without programming

Available Blocks

1. LLM Block

What it does: Communicates with AI language models to generate responses based on your prompts.

Use cases:

  • Generate creative content (stories, poems, articles)
  • Answer questions on any topic
  • Transform or reformat text
  • Analyze and summarize information

Key features:

  • Customizable AI model selection
  • Adjustable response creativity (temperature)
  • Support for multi-turn conversations
  • Automatic retry on connection failures

2. LLM Chat Block

What it does: Specialized block for building conversational experiences with memory of previous messages.

Use cases:

  • Build customer support chatbots
  • Create interactive tutoring systems
  • Develop conversational assistants
  • Design dialogue-based applications

Key features:

  • Maintains conversation history
  • Supports system instructions for AI behavior
  • Stream responses in real-time
  • Configurable timeout and retry settings

3. Messages Generator Block

What it does: Prepares and formats conversation messages for AI models using templates.

Use cases:

  • Create dynamic prompts with variable inputs
  • Format multi-turn conversations
  • Build reusable prompt templates
  • Prepare structured data for AI processing

Key features:

  • Template-based message creation
  • Support for system, user, and assistant roles
  • Variable substitution (e.g., {{input}})
  • Combines existing conversations with new messages

Common Use Cases

💬 Customer Support Bot

Connect the LLM Chat block to handle customer inquiries with context-aware responses.

🌐 Translation Service

Use the LLM block with translation prompts to convert text between languages.

✍️ Content Generation

Combine the Messages Generator and LLM blocks to create articles, emails, or social media posts.

📊 Data Analysis Assistant

Feed data into the LLM block to get summaries, insights, and recommendations.

Getting Started

  1. Install the Package: Import this LLM package into your OOMOL platform
  2. Choose Your Block: Select the block that fits your use case
  3. Configure Settings:
    • Select your preferred AI model
    • Adjust response parameters (temperature, max tokens)
    • Set up your prompt template
  4. Connect Your Flow: Link blocks together to create your workflow
  5. Test & Run: Execute your flow and see the AI in action!

Configuration Guide

Model Selection

Choose from various AI models based on your needs:

  • deepseek-chat (default): Fast and cost-effective
  • Other supported models available in the model selector

Response Control

  • Temperature (0-1): Controls creativity (0 = focused, 1 = creative)
  • Top P (0-1): Controls response diversity
  • Max Tokens: Limits response length

Retry & Timeout

  • Timeout: Maximum wait time for responses (default: 30 seconds)
  • Retry Times: Number of automatic retries on failure (default: 0)
  • Retry Sleep: Delay between retries (default: 3.5 seconds)

Example Workflows

This project includes sample workflows in the flows/ directory:

  • chat: Interactive chatbot example
  • translate: Language translation workflow
  • history: Conversation with context memory

Support & Resources

Version

Current version: 0.3.5

License

Check the repository for license information.


Ready to build your first AI workflow? Import this package into OOMOL and start connecting blocks!

About

Builtin LLM blocks in OOMOL Studio

Resources

Stars

Watchers

Forks

Packages

No packages published

Contributors 5