Skip to main content
The Agent API supports multiple AI agents optimized for different use cases. Each agent has distinct performance characteristics and intelligence levels.

​
Gemini

Recommended for most use cases Gemini Computer Use provides fast, high-quality responses through our API. This agent delivers excellent intelligence and is optimized for quick task completion with reliable results. Characteristics:
  • Fast response times
  • Strong reasoning capabilities
  • Recommended default choice
  • Reliable performance across varied tasks
Example usage: 
curl -X POST "https://tron.tera.space/v1/task/create" \
  -H "Authorization: Bearer $API_KEY" \
  -H "content-type: application/json" \
  -d '{"agent":"gemini","prompt":"What does example.com homepage say?","mode":"text","stepLimit":10}'

​
Glitter

Limited production use Glitter is an experimental agent made with open source models. Glitter has different performance characteristics that make it ineffective in most cases. This agent has longer response times and may not be suitable for time-sensitive applications or complex tasks requiring high intelligence. Characteristics:
  • Slower response times
  • Limited capability
  • Not recommended for production workflows
  • Consider Gemini for better performance
When to consider:
  • Experimental testing
  • Non-critical workloads
  • Scenarios with generous timeout windows

​
Choosing an Agent

For most applications, Gemini is the recommended choice. It provides the best balance of speed, reliability, and intelligence. Use Glitter only if you have specific experimental requirements or are evaluating alternative models for specialized use cases.