Apollo Language Model
Developed over the past seven years in collaboration with 60,000 human agents, Apollo is a breakthrough language model for objective-oriented conversational agents. Apollo replaces the transformer architecture of traditional LLMs with a neuro-symbolic architecture that couples the conversational skills of generative AI with the predictability and actionability of rule-based AI, making it better suited for agents.
Companies of any kind can fine-tune the Apollo base model on a specific task and deploy conversational agents that work on their behalf. These fine-tuned agents can continuously improve from human feedback, which is integrated into their updated versions. They further feature fine-grained control, the capacity to adhere to the deploying company's policies, and a white-box view of their operations.
With native tool use, defined scope and guardrails, and superior steerability, Apollo Agents outclass LLM Agents across all conversational use cases.
Apollo does not train on company data; instead, it can be pre-integrated with any number of internal systems and seamlessly call the relevant API endpoint whenever necessary. Like transformer-based LLMs, Apollo has a "system prompt," broken down to structured, conditional policies.
Workflows and tools are native to Apollo and do not rely on function calling or chaining. This enables a 100% success rate in executing complex API calls and the ability to return grounded answers in each interaction from any number of endpoints or sources.
Unlike transformer-based LLMs, Apollo is fine-tuned at the interaction level. In the Simulator, companies can generate automated conversations simulating the agent in production in front of users. They then proceed to fine-tune the agent by correcting its output in each unsatisfactory interaction. This process can be repeated continuously to improve the performance and alignment of the agent.
Unlocked with Apollo are some exciting new agentic use cases, including the ability for companies to responsibly communicate with their customers using conversational AI and to sell their products and services without the need for human involvement, the example featured in the Live Demo Agent.
