WebsiteLegalPlatform

Models

Lleverage provides a curated selection of state-of-the-art AI models from leading providers. These models are carefully selected to ensure reliability, security, and optimal performance for various use cases.

Key Features

  • Curated selection of best-in-class AI models

  • Models from leading providers (OpenAI, Anthropic, Google, Together AI, Mistral)

  • Configurable model settings for fine-tuning outputs

  • AI-powered model selection recommendations

  • Custom configurations for different workflow needs

  • Secure API key management

How to Choose the Right Model

  1. Consider your workflow's specific requirements

  2. Use the Lleverage Model Advisor for recommendations based on your use case

  3. For simple tasks requiring speed, select faster models like GPT-4.1-nano or Gemini 2.5 Flash

  4. For complex reasoning tasks, select more powerful models like GPT-4.1 or Claude 3.7 Sonnet

  5. For a balance of performance and cost, consider models like Llama 4 Scout or Mistral Large 2

💡 The Lleverage Model Advisor analyzes your prompt and workflow to recommend the optimal model for your specific use case.

How to Configure Model Settings

  1. Access model configuration in your workflow

  2. Select your preferred model provider and specific model

  3. Adjust the following parameters as needed:

    • Max Tokens: Controls response length (-1 for dynamic calculation)

    • Temperature: Controls creativity and randomness

    • Top-P (Nucleus): Manages probability threshold for token selection

    • Frequency Penalty: Reduces repetition in generated text

    • Presence Penalty: Enhances variety in responses

    • Stop Sequences: Defines when the model should stop generating

💡 For beginners, use the AI helper feature to automatically configure these settings based on your prompt.

How to Understand Temperature Settings

  1. Low temperature (close to 0):

    • More deterministic, predictable responses

    • Better for factual tasks, document analysis, data extraction

    • More focused and coherent outputs

    • Less creative variety

  2. High temperature (close to 1):

    • More creative, diverse responses

    • Better for brainstorming, creative writing, idea generation

    • More varied and potentially surprising outputs

    • Higher risk of tangential or less relevant content

⚠️ Higher temperature settings can produce more creative outputs but may occasionally generate less coherent or relevant responses.

How to Add API Keys to Models

  1. Navigate to Connections → Models in your Lleverage account

  2. Select the model provider you wish to configure

  3. Enter your API key from the corresponding provider

  4. Save your changes

  5. Your workflows can now use these models with your credentials

⚠️ Keep your API keys secure and never share them. Lleverage safely stores these keys for your workflows.

Available Model Providers and Models

OpenAI

  • GPT-4.1

  • GPT-4.1-mini

  • GPT-4.1-nano

  • GPT-o4-mini

Anthropic

  • Claude 3.7 Sonnet

Google

  • Gemini 2.5 Pro

  • Gemini 2.5 Flash

  • Gemini 2.0 Flash

Together AI

  • Llama 4 Maverick

  • Llama 4 Scout

  • DeepSeek V3

  • DeepSeek R1

Mistral

  • Mistral Large 2

💡 This list represents the most current models available on Lleverage. Older models remain accessible but are delisted from the primary selection.

Best Practices for Model Selection

  1. Test multiple models on your specific use case to compare performance

  2. Consider the balance between model capability and execution speed

  3. Start with the Model Advisor recommendation and fine-tune from there

  4. Use more powerful models for complex reasoning, summarization, or creative tasks

  5. Use faster models for simple tasks, quick responses, or high-volume workflows

  6. Create and save multiple model configurations for different workflow types

💡 Lleverage continuously updates its model offerings to ensure you have access to the most reliable and effective AI models available.

Last updated