Model Selection

Choose which Large Language Model (LLM) powers your content generation using the model parameter.

Each model offers different trade-offs between quality and speed. Currently, cost is the same regardless, though this may change in the future.

Model Options:

• opus: Most powerful model, excels at reasoning and complex instructions.

  • Best for: Highly structured, multi-step content with deep reasoning
  • Highest quality, but slowest response time
  • Email generation: ~5-7 seconds
  • Sequence generation (e.g., 4 emails): ~25-35 seconds

• sonnet_3.5 (v2): Balanced performance model, optimized for conversational and structured content.

  • Best for:
    • General-purpose content
    • Multi-email sequences (more conversational tone)
  • Default for: contentType: "sequence"
  • Email generation: ~4-6 seconds
  • Sequence generation (e.g., 4 emails): ~20-30 seconds

• gemini: Google's Gemini 1.5 Pro – excellent for long-form content with strong analytical capabilities.

  • Best for: Analytical content, structured writing, deep context handling
  • Default for:
    • contentType: "email"
    • contentType: "callScript"
    • contentType: "custom"
    • contentType: "opener"
    • contentType: "sms"
    • contentType: "linkedinConnectionRequest"
  • Email generation: ~3-5 seconds

• gemini_flash_1.5: Google's Gemini 1.5 Flash – optimized for extreme speed.

  • Best for: Real-time generation where speed is the priority
  • Ideal for:
    • Chat-based interactions
    • Rapid iteration
    • Large-scale content generation
  • Email generation: ~2-3 seconds

• gpt4o: Optimized for speed, good for real-time applications.

  • Best for:
    • Fast, dynamic content
    • Website personalization
  • Email generation: ~2-5 seconds

How to use the model parameter

When making a content generation request against the generate-content endpoint, the model parameter can be explicitly set for added control.

If the model parameter is omitted, the system will pick one by default (gemini 1.5 pro for email, sonnet 3.5 v2 for sequence, etc). Below is an example request:

JSON

{
  "contactEmail": "[e-mail]",
  "userEmail": "[e-mail]",
  "contentType": "email",
  "model": "gemini"
}

Breakdown by contentType

Content Type	Available Models	Default Model
Sequence	opus, sonnet_3.5, gpt4o, gemini_1.5_pro, gemini_flash_1.5	sonnet_3.5
Email	opus, sonnet_3.5, gpt4o, gemini_pro_1.5, gemini_flash_1.5	gemini_pro_1.5
Custom	opus, sonnet_3.5, gpt4o, gemini_pro_1.5, gemini_flash_1.5	gpt4o
Opener	gemini_pro_1.5	gemini_pro_1.5
Call Script	gemini_pro_1.5	gemini_pro_1.5