Model Selection
Not all AI models are the same. Model selection matches the right capability to your specific task for optimal performance and cost.
The process of choosing the right AI model for a specific task based on factors like accuracy, speed, cost, and capability - ranging from fast lightweight models for simple classification to powerful frontier models for complex reasoning.
Task Matching
Simple tasks (classification, routing) use fast, cheap models. Complex tasks (reasoning, drafting) use powerful models.
Speed vs. Capability
Small models respond in milliseconds for chat. Large models take seconds but handle nuanced reasoning.
Cost Optimization
Using the right model size for each task keeps costs predictable without sacrificing quality.
Model Fallbacks
If one model is unavailable or slow, the agent can fall back to alternatives - maintaining uptime.
Performance Monitoring
Track accuracy, response times, and costs per model. Optimize selection based on real performance data.
Automatic Upgrades
As better models become available, your agent can be upgraded - often with improved performance at lower cost.
FAQ
Do I choose the model?
We recommend the optimal model during setup based on your use case. You can request changes if needed.
Can different tasks use different models?
Yes. Email classification might use a fast, cheap model while customer response drafting uses a more capable one.
What happens when new models are released?
We evaluate new models and recommend upgrades when they improve performance or reduce costs for your use case.
Get the right model for your needs
We match the optimal AI model to your use case.