Model Presets

Overview

Model Presets are curated configurations that bundle a transcriber, model, and voice for a specific goal. Instead of picking each component individually, you can start from a preset that matches how your assistant needs to behave, then customize from there.

Balanced is the default preset for all new assistants. It provides a strong mix of quality, responsiveness, and cost, making it the best starting point for most use cases such as support, scheduling, FAQs, or qualification.

High Intelligence prioritizes reasoning and conversation quality. Choose it when:

• Conversations are complex, open-ended, or high-stakes
• The assistant needs to reason through multi-step problems or use tools reliably
• Accuracy and quality matter more than speed or cost

More capable models respond a little slower and cost more per minute than Balanced.

Ultra Fast prioritizes low latency and responsive conversations. Choose it when:

• Responsiveness is the priority and replies should feel instant
• Calls are high-volume and relatively straightforward
• The flow is scripted or transactional rather than open-ended

The fastest models are smaller and less capable, so Ultra Fast may not suit complex reasoning or heavy tool use.

Cost Saver prioritizes the lowest cost per minute. Choose it when:

• Call volume is high and you're optimizing spend
• Interactions are simple or scripted
• A small quality trade-off is acceptable in exchange for lower cost

The lowest-cost models are less capable, so this preset fits simpler interactions best.

To apply a preset:

1. Open your assistant in the Sulus dashboard
2. Select an option under Model Presets
3. Optionally customize any individual component (transcriber, model, or voice)
4. Click Publish

Presets are a starting point, not a lock. If you swap out any individual component, your assistant moves to a Customized state, and its performance metrics update to reflect the new configuration.

In summary, Model Presets give you a fast way to align your assistant's transcriber, model, and voice with your priorities, whether that's raw intelligence, speed, or cost, and you can always fine-tune individual components afterward.