Agents

OpenAGI: When LLM Meets Domain Experts

First page

Paper summary

An open-source research platform for LLM agents manipulating domain expert models.

Ask this paper

Key points

LLM-as-orchestrator platform: LLMs plan and orchestrate calls to specialized domain expert models (vision, speech, language).

Multi-step task evaluation: Provides a standardized evaluation framework for complex multi-step tasks requiring tool composition.

Open research tooling: Fully open-source platform for academic researchers to compare agent designs and tool-use strategies.

Agent research infrastructure: Part of the 2023 wave establishing shared infrastructure for LLM agent research.

Every Monday