Models
A model in Orion101 refers to an AI program that has been trained on large datasets to perform tasks, recognize patterns, and make decisions. Models are the foundational building blocks for agents, workflows, and tools, allowing them to deliver intelligent and contextually relevant responses. Popular examples include OpenAI's GPT-4o and Anthropic's Claude 3.5 Sonnet.
Orion101 includes popular preconfigured models, such as those offered by OpenAI, to ensure quick and seamless integration. However, administrators have the flexibility to modify these models or add new ones based on their specific requirements. Agents, workflows, and tools can specify which model they should use for their tasks. If no model is explicitly selected, Orion101 defaults to the system’s preconfigured model.
Supported Model Providers
Orion101 supports multiple model providers to give users the flexibility to choose the AI technology that best suits their needs. Currently, the platform integrates with the following providers:
OpenAI
Azure OpenAI
Anthropic
Ollama
Customizing and Configuring Models
For users who wish to utilize models from providers other than OpenAI, the corresponding Model Provider must first be enabled. Each provider requires specific setup steps, such as API keys or custom configurations, to establish integration. Orion101 simplifies this process with detailed documentation and an easy-to-follow Model Provider Configuration Guide.
Orion101’s model system ensures flexibility and scalability, enabling users to leverage cutting-edge AI technologies while maintaining control over their configurations. Whether you are using preconfigured models or customizing your own, Orion101 offers a robust platform for deploying the intelligence required for your agents, workflows, and tools.
Last updated